YOSHIOKA Masaharu
Faculty of Information Science and Technology Computer Science and Information Technology Knowledge Software Science | Professor |
■Researcher basic information
Researchmap personal page
Home Page URL
J-Global ID
Research Keyword
Research Field
■Career
Career
- Apr. 2019 - Present
Hokkaido University, Faculty of Information Science and Technology Division of Computer Science and Information Technology, 教授 - Jan. 2018 - Mar. 2019
Hokkaido University, Graduate School of Information Science and Technology Division of Computer Science and Information Technology, Professor - Apr. 2014 - Dec. 2017
Hokkaido University, Graduate School of Information Science and Technology, Division of Computer Science and Information Technology, Associate Professor - Apr. 2008 - Mar. 2014
Hokkaido University, Graduate School of Information Science and Technology Division of Computer Science, 准教授 - Apr. 2004 - Mar. 2008
Hokkaido University, Graduate School of Information Science and Technology Division of Computer Science, Associate Professor - Jun. 2001 - Mar. 2004
Hokkaido University, Graduate School of Engineering, 助教授 - Apr. 2000 - May 2001
National Institute of Informatics, 助手 - Apr. 1996 - Mar. 2000
The National Center for Science Information Systems, Research Associate
Committee Memberships
- Jan. 2024 - Present
ACM SIGIR, Steering committee member of International Conference on the Theory of Information Retrieval (ICTIR), Society - Dec. 2017 - Jun. 2021
Tokyo ACM SIGIR Chapter, Treasurer, Society - Jun. 2018 - Jun. 2020
人工知能学会, 理事, Society - 2009 - 2019
Steering committee member of Asian Information Retrieval Societies, Society - Apr. 2013 - Mar. 2015
情報処理学会, 情報基礎とアクセス技術研究会 幹事, Society - 2011 - 2013
情報処理学会 情報基礎とアクセス技術研究会 運営委員, Society - 2009 - 2011
情報処理学会, 情報学基礎研究会 運営委員, Society - 2009 - 2011
情報処理学会, 研究会論文誌TOD編集委員, Society - 2009
日本機械学会, 設計工学システム部門 代議員, Society - 2007 - 2008
情報処理学会, 情報学基礎研究会 幹事, Society - 2007 - 2008
情報処理学会, 研究会論文誌TOD共同編集委員長, Society - 2006 - 2007
日本機械学会, 設計工学システム部門 広報委員会幹事, Society - 2006 - 2007
人工知能学会, 編集委員, Society - 2005 - 2006
情報処理学会, 情報学基礎研究会 主査, Society - 2003 - 2006
情報処理学会, 研究会論文誌TOD編集委員, Society - 2004 - 2005
人工知能学会, 人工知能基本問題研究会 幹事, Society - 2001 - 2005
人工知能学会, 編集委員, Society - 2004
情報処理学会, 情報学基礎研究会 幹事, Society - 2003
情報処理学会, 情報学基礎研究会 運営委員, Society
■Research activity information
Awards
- Feb. 2022, Linked Open Data チャレンジ 2021 実行委員会, オントロジー賞(オントロノミー合同会社)
Wikipediaカテゴリオントロジー
北海道大学大学院情報科学研究院知識ベース研究室 - Feb. 2022, Linked Open Data チャレンジ 2021 実行委員会, データ作成部門 優秀賞
Wikipediaカテゴリオントロジー
北海道大学大学院情報科学研究院知識ベース研究室 - Jun. 2015, 人工知能学会, 人工知能学会 研究会優秀賞
複数国の新聞からの多観点比較による分析~GDELTデータを用いた分析~
吉岡真治;神門典子, Japan society
Papers
- Preliminary Experiments of Qualitative Reasoning Model Construction Using Large Language Model
Shinpei Suzuki, Masaharu Yoshioka
Proceedings of the QR 2024 - 37th International Workshop on Qualitative Reasoning, 58, 63, Oct. 2024, [Peer-reviewed], [Last author]
English, International conference proceedings - Overview of Benchmark Datasets and Methods for the Legal Information Extraction/Entailment Competition (COLIEE) 2024.
Randy Goebel, Yoshinobu Kano, Mi-Young Kim, Juliano Rabelo, Ken Satoh, Masaharu Yoshioka
JSAI-isAI, 109, 124, May 2024, [Peer-reviewed]
International conference proceedings - Overview and Discussion of the Competition on Legal Information, Extraction/Entailment (COLIEE) 2023.
Randy Goebel, Yoshinobu Kano, Mi-Young Kim, Juliano Rabelo, Ken Satoh, Masaharu Yoshioka
Rev. Socionetwork Strateg., 18, 1, 27, 47, Apr. 2024, [Peer-reviewed]
Scientific journal - Coding Open-Ended Responses using Pseudo Response Generation by Large Language Models.
Yuki Zenimoto, Ryo Hasegawa, Takehito Utsuro, Masaharu Yoshioka, Noriko Kando
NAACL (Student Research Workshop), 242, 254, 2024
International conference proceedings - OSPAR: A Corpus for Extraction of Organic Synthesis Procedures with Argument Roles.
Kojiro Machi, Seiji Akiyama, Yuuya Nagata, Masaharu Yoshioka
Journal of chemical information and modeling, 63, 21, 6619, 6628, 19 Oct. 2023, [Peer-reviewed], [Last author, Corresponding author], [International Magazine]
English, Scientific journal, There is a pressing need for the automated extraction of chemical reaction information because of the rapid growth of scientific documents. The previously reported works in the literature for the procedure extraction either (a) did not consider the semantic relations between the action and argument or (b) defined a detailed schema for the extraction. The former method was insufficient for reproducing the reaction, while the latter methods were too specific to their own schema and did not consider the general semantic relation between the verb and argument. In addition, they did not provide an annotated text that aligned with the structured procedure. Along these lines, in this work, we propose a corpus named organic synthesis procedures with argument roles (OSPAR) that is annotated with rolesets to consider the semantic relation between the verb and argument. We also provide rolesets for chemical reactions, especially for organic synthesis, which represent the argument roles of actions in the corpus. More specifically, we annotated 112 organic synthesis procedures in journal articles from Organic Syntheses and defined 19 new rolesets in addition to 29 rolesets from an existing language resource (Proposition Bank). After that, we constructed a simple deep learning system trained on OSPAR and discussed the usefulness of the corpus by comparing it with chemical description language (XDL) generated by a natural language processing tool, namely, SynthReader. While our system's output required more detailed parsing, it covered comparable information against XDL. Moreover, we confirmed that the validation of the output action sequence was easy as it was aligned with the original text. - COLIEE 2022 Summary: Methods for Legal Document Retrieval and Entailment
Mi-Young Kim, Juliano Rabelo, Randy Goebel, Masaharu Yoshioka, Yoshinobu Kano, Ken Satoh
New Frontiers in Artificial Intelligence, 51, 67, Springer Nature Switzerland, 06 Apr. 2023, [Peer-reviewed]
In book - Summary of the Competition on Legal Information, Extraction/Entailment (COLIEE) 2023.
Randy Goebel, Yoshinobu Kano, Mi-Young Kim, Juliano Rabelo 0001, Ken Satoh, Masaharu Yoshioka
ICAIL, 472, 480, 2023, [Peer-reviewed]
International conference proceedings - Data-Augmentation Method for BERT-based Legal Textual Entailment Systems in COLIEE Statute Law Task
Yasuhiro Aoki, Masaharu Yoshioka, Youta Suzuki
REVIEW OF SOCIONETWORK STRATEGIES, 16, 175, 196, SPRINGER JAPAN KK, Feb. 2022, [Peer-reviewed], [Corresponding author]
English, Scientific journal, A legal textual entailment task is a task to recognize entailment between a law article and its statements. In the Competition on Legal Information Extraction/Entailment (COLIEE), this task is designed as a task to confirm the entailment of a yes/no answer from the given civil code article(s). Based on the development of deep-learning-based natural language processing tools such as bidirectional encoder representations from transformers (BERT), many participants in the task used such tools, and the best performance system of COLIEE 2020 was a BERT-based system. However, because of the limitation of the size of training data provided by the task organizer, training such tools to adapt to the variability of the questions is difficult. In this paper, we propose a data-augmentation method to make training data using civil code articles for understanding the syntactic structure of the questions and articles for entailment. Our BERT-based ensemble system, which uses this augmentation method, achieves the best performance (accuracy = 0.7037) in Task 4 of COLIEE 2021. We also introduce the results of additional experiments to discuss the characteristics of the proposed method. - Overview and Discussion of the Competition on Legal Information Extraction/Entailment (COLIEE) 2021
Juliano Rabelo, Randy Goebel, Mi-Young Kim, Yoshinobu Kano, Masaharu Yoshioka, Ken Satoh
REVIEW OF SOCIONETWORK STRATEGIES, 18, 1, 27, 47, SPRINGER JAPAN KK, Feb. 2022, [Peer-reviewed]
English, Scientific journal, We summarize the 8th Competition on Legal Information Extraction and Entailment. In this edition, the competition included five tasks on case law and statute law. The case law component includes an information retrieval Task (Task 1), and the confirmation of an entailment relation between an existing case and an unseen case (Task 2). The statute law component includes an information retrieval Task (Task 3), an entailment/question answering task based on retrieved civil code statutes (Task 4) and an entailment/question answering task without retrieved civil code statutes (Task 5). Participation was open to any group based on any approach. Eight different teams participated in the case law competition tasks, most of them in more than one task. We received results from six teams for Task 1 (16 runs) and 6 teams for Task 2 (17 runs). On the statute law task, there were eight different teams participating, most in more than one task. Six teams submitted a total of 18 runs for Task 3, 6 teams submitted a total of 18 runs for Task 4, and 4 teams submitted a total of 12 runs for Task 5. Here we summarize the approaches, our official evaluation, and analysis on our data and submission results. - HUKB at the COLIEE 2022 Statute Law Task.
Masaharu Yoshioka, Youta Suzuki, Yasuhiro Aoki
New Frontiers in Artificial Intelligence - JSAI-isAI 2022 Workshop, JURISIN 2022, and JSAI 2022 International Session, 109, 124, Springer, 2022, [Peer-reviewed], [Lead author]
International conference proceedings - HUKB at ChEMU 2022 Task 1: Expression-Level Information Extraction.
Kojiro Machi, Masaharu Yoshioka
CLEF (Working Notes), 797, 807, CEUR-WS.org, 2022
International conference proceedings - Interactive Visualization of the Polarity-based Stance of News Websites using News Genres
Masaharu Yoshioka, Norihiko Tatsunami, Masahiko Itoh, Noriko Kando, James Allan
Proceedings of the 9th International Workshop on News Recommendation and Analytics (INRA 2021), Sep. 2021, [Peer-reviewed], [Lead author]
English, International conference proceedings - Proposal of the Aesthetic Experience-Oriented Evaluation Framework for Field-recording Sound Retrieval System: Experiments using Acoustic Feature Signatures Based on Multiscale Fractal Dimension.
Motohiro Sunouchi, Masaharu Yoshioka
IVSP 2021: 3rd International Conference on Image, Video and Signal Processing(IVSP), 65, 72, ACM, Mar. 2021, [Peer-reviewed], [Last author]
English, International conference proceedings - COLIEE 2020: Methods for Legal Document Retrieval and Entailment
Juliano Rabelo, Mi Young Kim, Randy Goebel, Masaharu Yoshioka, Yoshinobu Kano, Ken Satoh
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 12758 LNAI, 196, 210, 2021, [Peer-reviewed], [Internationally co-authored]
English, International conference proceedings, We present a summary of the 7th Competition on Legal Information Extraction and Entailment. The competition consists of four tasks on case law and statute law. The case law component includes an information retrieval task (Task 1), and the confirmation of an entailment relation between an existing case and an unseen case (Task 2). The statute law component includes an information retrieval task (Task 3) and an entailment/question answering task (Task 4). Participation was open to any group based on any approach. Ten different teams participated in the case law competition tasks, most of them in more than one task. We received results from 9 teams for Task 1 (22 runs) and 8 teams for Task 2 (22 runs). On the statute law task, there were 14 different teams participating, most in more than one task. Eleven teams submitted a total of 28 runs for Task 3, and 13 teams submitted a total of 30 runs for Task 4. We summarize the approaches, our official evaluation, and analysis on our data and submission results. - Diversity-robust acoustic feature signatures based on multiscale fractal dimension for similarity search of environmental sounds
Motohiro SUNOUCHI, Masaharu YOSHIOKA
IEICE Transactions on Information and Systems, E104D, 10, 1734, 1748, 2021, [Peer-reviewed], [Last author]
Scientific journal, This paper proposes new acoustic feature signatures based on the multiscale fractal dimension (MFD), which are robust against the diversity of environmental sounds, for the content-based similarity search. The diversity of sound sources and acoustic compositions is a typical feature of environmental sounds. Several acoustic features have been proposed for environmental sounds. Among them is the widely-used Mel-Frequency Cepstral Coefficients (MFCCs), which describes frequency-domain features. However, in addition to these features in the frequency domain, environmental sounds have other important features in the time domain with various time scales. In our previous paper, we proposed enhanced multiscale fractal dimension signature (EMFD) for environmental sounds. This paper extends EMFD by using the kernel density estimation method, which results in better performance of the similarity search tasks. Furthermore, it newly proposes another acoustic feature signature based on MFD, namely very-long-range multiscale fractal dimension signature (MFD-VL). The MFD-VL signature describes several features of the time-varying envelope for long periods of time. The MFD-VL signature has stability and robustness against background noise and small fluctuations in the parameters of sound sources, which are produced in field recordings. We discuss the effectiveness of these signatures in the similarity sound search by comparing with acoustic features proposed in the DCASE 2018 challenges. Due to the unique descriptiveness of our proposed signatures, we confirmed the signatures are effective when they are used with other acoustic features. - BERT-based ensemble methods with data augmentation for legal textual entailment in COLIEE statute law task.
Masaharu Yoshioka, Yasuhiro Aoki, Youta Suzuki
Proceedings of the 18th International Conference on Artificial Intelligence and Law, ICAIL 2021, 278, 284, ACM, 2021, [Peer-reviewed], [Lead author]
International conference proceedings, The Competition on Legal Information Extraction/Entailment (COLIEE) statute law legal textual entailment task (task 4) is a task to make a system judge whether a given question statement is true or not by provided articles. In the last COLIEE 2020, the best performance system used bidirectional encoder representations from transformers (BERT), a deep-learning-based natural language processing tool for handling word semantics by considering their context. However, there are problems related to the small amount of training data and the variability of the questions. In this paper, we propose a BERT-based ensemble method with data augmentation to solve this problem. For the data augmentation, we propose a systematic method to make training data for understanding the syntactic structure of the questions and articles for entailment. In addition, due to the nature of the non-deterministic characteristics of BERT fine-tuning and the variability of the questions, we propose a method to construct multiple BERT fine-tuning models and select an appropriate set of models for ensemble. The accuracy of our proposed method for task 4 was 0.7037, which was the best performance among all submissions. - HUKB at ChEMU 2021 Task 2: Anaphora Resolution.
Kojiro Machi, Masaharu Yoshioka
CEUR Workshop Proceedings, 2936, 720, 731, CEUR-WS.org, 2021, [Last author]
English, International conference proceedings, This paper describes our system for the ChEMU 2021 task 2 of anaphora resolution for extracting chemical reactions from patents. We divide the task into two subtasks: (1) span detection of mentions that include antecedent and anaphor; and (2) mention classification and relation detection. For the first task, we use a deep learning-based approach. For the second task, we use a rule-based approach that uses features related to chemical reactions by ChemicalTagger, which is a state-of-the-art text-mining tool for chemistry. Our system obtained an exact-match F-score of 0.6907 and a relaxed match F-score of 0.7459. - Temporal Information Access
Masaharu Yoshioka, Hideo Joho
Information Retrieval Series, 43, 127, 141, 2021, [Invited], [Lead author]
English, In book, This chapter introduces the research background and details of temporal information access tasks in the NTCIR. The GeoTime task was the first attempt to evaluate temporal information retrieval as an extension of an information-retrieval-for-question-answering task. Temporalia was a task to investigate the role of temporal factors in a search. - Wikipedia Category Ontology: A Framework for Utilization of the Wikipedia Category Structure by Knowledge Engineers.
Masaharu Yoshioka, Takanori Nakagawa
CEUR Workshop Proceedings, 2721, 300, 305, CEUR-WS.org, 2020, [Peer-reviewed], [Lead author]
English, International conference proceedings, Wikipedia categories are intended to group together pages on similar topics and are organized in a hierarchical structure. Since the editorial policy for Wikipedia categories differs from policies used by knowledge engineers, various types of relationship exist in its category structure. In this paper, we propose a novel framework called "Wikipedia category ontology" (WCO) that aims to act as a basis for interpreting the Wikipedia category structure and is based on a classification of category and relationship types. WCO enables particular Wikipedia category substructures to be extracted, including class-subclass hierarchies, class-instance, and sets of diffused categories for a particular category. WCO is available online in the form of linked open data at http://wcontology.org/. - COLIEE : Competition for Legal Information Extraction and Entailment
狩野芳伸, 吉岡真治, RABELO Juliano, KIM Mi-Young, GOEBEL Randy, 佐藤健
Journal of the Japanese Society for Artificial Intelligence, 35, 3, 377, 384, The Japanese Society for Artificial Intelligence, 2020, [Invited], [Internationally co-authored], [Domestic magazines]
Japanese - A Summary of the COLIEE 2019 Competition.
Juliano Rabelo, Mi-Young Kim, Randy Goebel, Masaharu Yoshioka, Yoshinobu Kano, Ken Satoh
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 12331 LNAI, 34, 49, Springer, 2019, [Peer-reviewed], [Internationally co-authored]
International conference proceedings, We summarize the evaluation of the 6th Competition on Legal Information Extraction/Entailment (COLIEE 2019). The competition consists of four tasks: two on case law and two on statute law. The case law component includes an information retrieval task (Task 1), and the confirmation of an entailment relation between an existing case and an unseen case (Task 2). The statute law component also includes an information retrieval task (Task 3) and an entailment/question answering task (Task 4), which attempts to confirm whether a particular statute applies to a yes/no question. Participation was open to any group in the world, based on any approach. Eleven different teams participated in the case law competition tasks, some of them in more than one task. We received results from 7 teams for Task 1 (15 runs) and 7 teams for Task 2 (18 runs). For the statute law tasks, 8 different teams participated, some in more than one task. Seven teams submitted a total of 13 runs for Task 3, and 7 teams submitted a total of 15 runs for Task 4. Here we summarize each team’s approaches, our official evaluation, and some analysis of the variety of methods that produced the evaluation results. - Construction of In-House Papers/Figures Database System using PDFs
M. Yoshioka, S. Hara
Collected Abstract of the 12th International Workshop on Information Search, Integration, and Personalization (ISIP 2018), Fukuoka, Japan, May 14-15, 2018., 1, 1, May 2018, [Peer-reviewed]
English - COLIEE-2018: Evaluation of the Competition on Legal Information Extraction and Entailment.
Yoshinobu Kano, Mi-Young Kim, Masaharu Yoshioka, Yao Lu, Juliano Rabelo, Naoki Kiyota, Randy Goebel, Ken Satoh
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 11717 LNAI, 177, 192, Springer, 2018, [Peer-reviewed], [Internationally co-authored]
International conference proceedings, We summarize the evaluation of the 5th Competition on Legal Information Extraction/Entailment 2018 (COLIEE-2018). The COLIEE-2018 tasks include two tasks in each of statute law and case law. The case law component includes an information retrieval (Task 1), and the confirmation of an entailment relation between an existing case and an unseen case (Task 2). The statute law component includes information retrieval (Task 3) and entailment/question answering (Task 4). Participation was open to any group based on any approach. 13 teams participated in the case law competition, and we received results from 7 teams where 6 submissions to Task 1 (12 runs), and 4 submissions to Task 2 (8 runs). Regarding the statute law, there were submissions of 17 runs from 8 teams (including 2 organizers’ runs) for Task 3 and 7 runs from 3 teams for Task 4. We describe each team’s approaches, our official evaluation, and analysis on our data and submission results. We also discuss possibilities for future competition tasks. - Construction of an In-House Paper/Figure Database System Using Portable Document Format Files.
Masaharu Yoshioka, Shinjiro Hara
Information Search, Integration, and Personalization (Communications in Computer and Information Science (CCIS), edited by D. Kotzinos, D. Laurent, N. Spyratos, Y. Tanaka, and R. Taniguchi, Springer International Publishing, Switzerland, 2019), Vol. 1040, 41, 52, Springer, 2018, [Peer-reviewed], [Lead author]
English, International conference proceedings, Several general-purpose databases of research papers are available, including ScienceDirect and Google Scholar. However, these systems may not include recent papers from workshops or conferences for special interest groups. Moreover, it can be helpful for researchers in a particular domain to analyze research papers using their own terminology. To support these researchers, we propose a new database system based on the information extracted from portable document format documents that enables annotation of terms via a terminology extraction system. In this paper, we evaluate our system using a use-case experiment and discuss the appropriateness of the system. - Visualizing Polarity-based Stances of News Websites.
Masaharu Yoshioka, Myungha Jang, James Allan, Noriko Kando
Proceedings of the Second International Workshop on Recent Trends in News Information Retrieval co-located with 40th European Conference on Information Retrieval (ECIR 2018), Grenoble, France, March 26, 2018., 2079, 6, 8, CEUR-WS.org, 2018, [Peer-reviewed], [Lead author], [Internationally co-authored]
International conference proceedings, We develop a novel framework that helps identify potential bias in news websites to support users who are exposed to news articles with a wide variety of political leanings. We propose a polarity-based stance (PS), a vector that represents how often a website publishes articles that are positive or negative with regard to a topic. We derive PS using the GDELT database and visualize the news websites' stances. We demonstrate the utility of our framework via a case study of the 2016 US Presidential Election. - Analysis of COLIEE Information Retrieval Task Data
Masaharu Yoshioka
NEW FRONTIERS IN ARTIFICIAL INTELLIGENCE (JSAI-ISAI 2017), 10838, 5, 19, SPRINGER INTERNATIONAL PUBLISHING AG, 2018, [Peer-reviewed], [Lead author]
English, International conference proceedings, The Competition on Legal Information Extraction/Entailment (COLIEE) involves the legal question answering task. The information retrieval task for finding relevant articles to questions is one of the subtasks in COLIEE. In this paper, we compare the characteristics of the test data provided in two different language (English and Japanese) and analyze topic difficulty based on the submission data by using the retrieval results of Indri, a state-of-the-art information retrieval system. We also discuss issues relating to the design of new COLIEE information retrieval tasks. - Measuring Beginner Friendliness of Japanese Web Pages explaining Academic Concepts by Integrating Neural Image Feature and Text Features
Hayato Shiokawa, Kota Kawaguchi, Bingcai Han, Takehito Utsuro, Yasuhide Kawada, Masaharu Yoshioka, Noriko Kando
NATURAL LANGUAGE PROCESSING TECHNIQUES FOR EDUCATIONAL APPLICATIONS, 143, 151, ASSOC COMPUTATIONAL LINGUISTICS-ACL, 2018, [Peer-reviewed]
English, International conference proceedings, Search engine is an important tool of modern academic study, but the results are lack of measurement of beginner friendliness. In order to improve the efficiency of using search engine for academic study, it is necessary to invent a technique of measuring the beginner friendliness of a Web page explaining academic concepts and to build an automatic measurement system. This paper studies how to integrate heterogeneous features such as a neural image feature generated from the image of the Web page by a variant of CNN (convolutional neural network) as well as text features extracted from the body text of the HTML file of the Web page. Integration is performed through the framework of the SVM classifier learning. Evaluation results show that heterogeneous features perform better than each individual type of features. - WC3: Wikipedia Category Comprehensiveness Checker based on the DBpedia Metadata Database.
Masaharu Yoshioka
Workshop and Poster Proceedings of the 8th Joint International Semantic Technology Conference co-located with the 8th Joint International Semantic Technology Conference (JIST 2018), Awaji City, Hyogo, Japan, November 26-28, 2018., 2293, 98, 101, CEUR-WS.org, 2018, [Peer-reviewed], [Lead author]
International conference proceedings, We demonstrate Wikipedia Category Comprehensiveness Checker (WC3) based on the DBpedia metadata database. This system supports to check comprehensiveness and consistency of Wikipedia category annotation using DBpedia database. This system is available online. - Construction of In-House Papers/Figures Database System for a Paticular Research Domain using PDFs - Application of Nano-Crystal Device Development Domain -
M. Yoshioka, S. Hara
Proccedings of the 2nd International Workshop on Scientific Document Analysis (SCIDOCA 2017), Tokyo, Japan, November 14-15, 2017: Proccedings of the 2nd International Workshop on Scientific Document Analysis (7pp.), 1, 7, Nov. 2017, [Peer-reviewed]
English - WC3: Analyzing the Style of Metadata Annotation Among Wikipedia Articles by Using Wikipedia Category and the DBpedia Metadata Database
Masaharu Yoshioka
KNOWLEDGE GRAPHS AND LANGUAGE TECHNOLOGY, 10579, 119, 136, SPRINGER INTERNATIONAL PUBLISHING AG, 2017, [Peer-reviewed], [Lead author]
English, International conference proceedings, WC3 (Wikipedia Category Consistency Checker) is a system that supports the analysis of the metadata-annotation style in Wikipedia articles belonging to a particular Wikipedia category (the subcategory of "Categories by parameter") by using the DBpedia metadata database. This system aims to construct an appropriate SPARQL query to represent the category and compares the retrieved results and articles that belong to the category. In this paper, we introduce WC3 and extend the algorithm to analyze efficiently additional varieties of Wikipedia category. We also discuss the metadata-annotation quality of the Wikipedia by usingWC3. URL ofWC3 is http://wnews.ist.hokudai.ac.jp/wc3/ and related files are available at http://wnews.ist.hokudai.ac.jp/wc3/KEKI. - A Civil Code Article Information Retrieval System based on Phrase Alignment with Article Structure Analysis and Ensemble Approach.
Masaharu Yoshioka, Daiki Onodera
COLIEE 2017. 4th Competition on Legal Information Extraction and Entailment, held in conjunction with the 16th International Conference on Artificial Intelligence and Law (ICAIL 2017) in King's College London, UK., 9, 22, EasyChair, 2017, [Peer-reviewed], [Lead author]
International conference proceedings - A multi-faceted figure retrieval system from research papers for supporting nano-crystal device development researchers
Masaharu Yoshioka, Tao Zhu, Shinjiroh Hara
The Proceedings of the First International Workshop on ScIentific Document Analysis (SCIDOCA 2016), Nov. 2016, [Peer-reviewed], [Lead author]
English, International conference proceedings - Civil Code Article Information Retrieval System based on Legal Terminology and Civil Code Article Structure
Daiki Onodera, Masaharu Yoshioka
The Proceedings of the 10th International Workshop on Juris-Informatics (JURISIN2016), The Japanese Society of Artificial Intelligence, Nov. 2016, [Peer-reviewed], [Lead author]
English, International conference proceedings - WC3: Analyzing the style of metadata annotation among Wikipedia articles by using Wikipedia category and the DBPedia metadata database.
Masaharu YOSHIOKA
Proceedings of the First Workshop on Knowledge Extraction and Knowledge Integration, Oct. 2016, [Peer-reviewed]
English, International conference proceedings - Interactive Metric Learning-Based Visual Data Exploration: Application to the Visualization of a Scientific Social Network
Masaharu Yoshioka, Masahiko Itoh, Michele Sebag
INFORMATION SEARCH, INTEGRATION, AND PERSONALIZATION, (ISIP 2015), 622, 142, 156, SPRINGER INTERNATIONAL PUBLISHING AG, 2016, [Peer-reviewed], [Lead author], [Internationally co-authored]
English, International conference proceedings, Data visualization is a core approach for understanding data specifics and extracting useful information in a simple and intuitive way. Visual data mining proceeds by projecting multidimensional data onto two-dimensional (2D) or three-dimensional (3D) data, e.g., through mathematical optimization and topology preserved in multidimensional scaling (MDS). However, this projection does not necessarily comply with the user's needs, prior knowledge and/or expectations. This paper proposes an interactive visual mining approach, centered on the user's needs and allowing the modification of data visualization by leveraging approaches from metric learning. The paper exemplifies the proposed system, referred to as Interactive Metric Learning-based Visual Data Exploration (IMViDE), applied to scientific social network browsing. - Comparative Analysis of GDELT Data Using the News Site Contrast System.
Masaharu Yoshioka, Noriko Kando
Proceedings of the First International Workshop on Recent Trends in News Information Retrieval co-located with 38th European Conference on Information Retrieval (ECIR 2016), Padua, Italy, March 20, 2016., 1568, 63, 65, CEUR-WS.org, 2016, [Peer-reviewed], [Lead author]
International conference proceedings, The News Site Contrast (NSContrast) system analyzes news articles retrieved from multiple news sites based on the concept of contrast set mining. It can extract terms that characterize different topics of interest across news sites, countries, and regions. In this study, we used NSContrast to analyze Global Database of Events, Language, and Tone (GDELT) data by comparing news articles from different regions (e.g., USA, Asia, and the Middle East). We also present examples of analyses performed using this system. - NaDev: An annotated corpus to support information extraction from research papers on nanocrystal devices
Thaer M. Dieb, Masaharu Yoshioka, Shinjiro Hara
Journal of Information Processing, 24, 3, 554, 564, Information Processing Society of Japan, 2016, [Peer-reviewed]
English, Scientific journal, The process of nanocrystal device development is not well systematized. To support this process, analysis of the information produced by developmental experiments is required. In this study, we constructed an annotated corpus to support the extraction of experimental information from relevant publications. We designed the corpus-construction guidelines by cooperating with a domain expert. We evaluated these guidelines through corpusconstruction experiments with graduate students from this domain, and then evaluated the corpus with the domain expert. In the corpus construction experiments, we achieved a sufficient level of Inter-Annotator Agreement by using a loose agreement measure that ignored the term-boundary mismatch problem, and made an agreement corpus that excluded annotations based on misunderstanding the guidelines. The domain expert evaluated this agreement corpus and modified the guidelines based on real examples. Using these guidelines, we finalized the corpus called "NaDev" (Nanocrystal Device development corpus). The NaDev corpus and its construction guidelines will be released via our website, http://nanoinfo.ist.hokudai.ac.jp/. The NaDev corpus aims to support automatic information extraction from publications relevant to nanocrystal device development. This information can be used to solve problems in the nanotechnology domain using the massive availability of fresh information. To the best of our knowledge, this is the first corpus constructed for the development of nanocrystal devices. - Extraction of Chemical and Drug Named Entities by Ensemble Learning Using Chemical NER Tools Based on Different Extraction Guidelines
Thaer M. Dieb, Masaharu Yoshioka
Transactions on Machine Learning and Data Mining, 8, 2, 61, 76, Oct. 2015, [Peer-reviewed], [Last author]
English, Scientific journal - Framework for automatic information extraction from research papers on nanocrystal devices
Thaer M. Dieb, Masaharu Yoshioka, Shinjiro Hara, Marcus C. Newton
BEILSTEIN JOURNAL OF NANOTECHNOLOGY, 6, 1872, 1882, BEILSTEIN-INSTITUT, Sep. 2015, [Peer-reviewed], [Internationally co-authored]
English, Scientific journal, To support nanocrystal device development, we have been working on a computational framework to utilize information in research papers on nanocrystal devices. We developed an annotated corpus called "NaDev" (Nanocrystal Device Development) for this purpose. We also proposed an automatic information extraction system called "NaDevEx" (Nanocrystal Device Automatic Information Extraction Framework). NaDevEx aims at extracting information from research papers on nanocrystal devices using the NaDev corpus and machine-learning techniques. However, the characteristics of NaDevEx were not examined in detail. In this paper, we conduct system evaluation experiments for NaDevEx using the NaDev corpus. We discuss three main issues: system performance, compared with human annotators; the effect of paper type (synthesis or characterization) on system performance; and the effects of domain knowledge features (e.g., a chemical named entity recognition system and list of names of physical quantities) on system performance. We found that overall system performance was 89% in precision and 69% in recall. If we consider identification of terms that intersect with correct terms for the same information category as the correct identification, i.e., loose agreement (in many cases, we can find that appropriate head nouns such as temperature or pressure loosely match between two terms), the overall performance is 95% in precision and 74% in recall. The system performance is almost comparable with results of human annotators for information categories with rich domain knowledge information (source material). However, for other information categories, given the relatively large number of terms that exist only in one paper, recall of individual information categories is not high (39-73%); however, precision is better (75-97%). The average performance for synthesis papers is better than that for characterization papers because of the lack of training examples for characterization papers. Based on these results, we discuss future research plans for improving the performance of the system. - Detecting bursty topics of correlated news and twitter for government services
Takehito Utsuro, Yusuke Inoue, Takakazu Imada, Masaharu Yoshioka, Noriko Kando
Social Media for Government Services, 135, 149, Springer International Publishing, 01 Jan. 2015, [Peer-reviewed]
English, In book, This chapter presents a framework of detecting bursty topics of correlated news and twitter, and discusses how to integrate the framework into government services. Especially, as a specific application of the proposed framework of detecting bursty topics of correlated news and twitter, this chapter gives an example of collecting news and twitter that are related to “the 2012 London Olympic game” and applying the proposed framework. - WC3:Wikipedia Category Consistency Checker based on DBPedia
Masaharu Yoshioka, Rhett Loban
2015 11TH INTERNATIONAL CONFERENCE ON SIGNAL-IMAGE TECHNOLOGY & INTERNET-BASED SYSTEMS (SITIS), 712, 718, IEEE, 2015, [Peer-reviewed], [Lead author]
English, International conference proceedings, Wikipedia is a free Internet encyclopedia that covers numerous topics and is maintained by voluntary editors. There have been several efforts to analyze the quality of Wikipedia's content, specifically assessing textual content and user editorial behavior. However, there has been no specific analysis of the quality of structured data (e.g., category structures, attributes contained in infoboxes) in Wikipedia. DBPedia is one of a core part of Linked Open Data database which extracts structured information from Wikipedia. Since the quality of DBPedia depends on Wikipedia, it is beneficial to have a mechanism to keep a consistent style of writing of the structured data (metadata) within Wikipedia. In this paper, we propose a system WC3 (WC-triple: Wikipedia Category Consistency Checker) that uses DBPedia RDF-triple information to evaluate the consistency of the metadata related to the Wikipedia category information. We also illustrate examples of typical inconsistencies that the system can assist in detecting. - The CHEMDNER corpus of chemicals and drugs and its annotation principles
Martin Krallinger, Obdulia Rabal, Florian Leitner, Miguel Vazquez, David Salgado, Zhiyong Lu, Robert Leaman, Yanan Lu, Donghong Ji, Daniel M. Lowe, Roger A. Sayle, Riza Theresa Batista-Navarro, Rafal Rak, Torsten Huber, Tim Rocktaschel, Serergio Matos, David Campos, Buzhou Tang, Hua Xu, Tsendsuren Munkhdalai, Keun Ho Ryu, S. V. Ramanan, Senthil Nathan, Slavko Zitnik, Marko Bajec, Lutz Weber, Matthias Irmer, Saber A. Akhondi, Jan A. Kors, Shuo Xu, Xin An, Utpal Kumar Sikdar, Asif Ekbal, Masaharu Yoshioka, Thaer M. Dieb, Miji Choi, Karin Verspoor, Madian Khabsa, C. Lee Giles, Hongfang Liu, Komandur Elayavilli Ravikumar, Andre Lamurias, Francisco M. Couto, Hong-Jie Dai, Richard Tzong-Han Tsai, Caglar Ata, Tolga Can, Anabel Usie, Rui Alves, Isabel Segura-Bedmar, Paloma Martinez, Julen Oyarzabal, Alfonso Valencia
JOURNAL OF CHEMINFORMATICS, 7, S-1, S2, BMC, Jan. 2015, [Peer-reviewed], [Internationally co-authored]
English, Scientific journal, The automatic extraction of chemical information from text requires the recognition of chemical entity mentions as one of its key steps. When developing supervised named entity recognition (NER) systems, the availability of a large, manually annotated text corpus is desirable. Furthermore, large corpora permit the robust evaluation and comparison of different approaches that detect chemicals in documents. We present the CHEMDNER corpus, a collection of 10,000 PubMed abstracts that contain a total of 84,355 chemical entity mentions labeled manually by expert chemistry literature curators, following annotation guidelines specifically defined for this task. The abstracts of the CHEMDNER corpus were selected to be representative for all major chemical disciplines. Each of the chemical entity mentions was manually labeled according to its structure-associated chemical entity mention (SACEM) class: abbreviation, family, formula, identifier, multiple, systematic and trivial. The difficulty and consistency of tagging chemicals in text was measured using an agreement study between annotators, obtaining a percentage agreement of 91. For a subset of the CHEMDNER corpus (the test set of 3,000 abstracts) we provide not only the Gold Standard manual annotations, but also mentions automatically detected by the 26 teams that participated in the BioCreative IV CHEMDNER chemical mention recognition task. In addition, we release the CHEMDNER silver standard corpus of automatically extracted mentions from 17,000 randomly selected PubMed abstracts. A version of the CHEMDNER corpus in the BioC format has been generated as well. We propose a standard for required minimum information about entity annotations for the construction of domain specific corpora on chemical and drug entities. The CHEMDNER corpus and annotation guidelines are available at: http://www.biocreative.org/resources/biocreative-iv/chemdner-corpus/ - Automatic Annotation of Parameters from Nanodevice Development Research Papers
T. M. Dieb, M. Yoshioka, S. Hara, M. C. Newton
Proceedings of the 4th International Workshop on Computational Terminology (COMPUTERM 2014), COLING 2014 Workshop, Dublin, Ireland, August 23, 2014: Proceedings of the 4th International Workshop on Computational Terminology, pp. 77-85 (9pp.), 77, 85, Aug. 2014, [Peer-reviewed], [Internationally co-authored]
English - Analysis of Japanese Wikipedia Category for Constructing Wikipedia Ontology and Semantic Similarity Measure
Masaharu Yoshioka
INFORMATION RETRIEVAL TECHNOLOGY, AIRS 2014, 8870, 470, 481, SPRINGER-VERLAG BERLIN, 2014, [Peer-reviewed], [Lead author]
English, International conference proceedings, Wikipedia is a free encyclopedia on the Internet that contains millions of articles. Wikipedia ontologies (e.g., YAGO2 and the Japanese Wikipedia ontology) have been constructed to utilize the semantic information in this encyclopedia; in these ontologies, the category information from each article is used to construct concept hierarchies. However, Wikipedia categories were originally designed for navigation, to help users find appropriate articles, and they are not equivalent to concept hierarchies in Wikipedia ontologies. In this study, we briefly review the definition of a category in Wikipedia and exhaustively investigate the category structure in the Japanese Wikipedia. We also discuss how category structure information can be utilized to construct an ontology and to calculate a semantic similarity measure. - Ensemble Approach to Extract Chemical Named Entity
Masaharu Yoshioka, Thaer M. Dieb
Proceedings of the fourth BioCreative challenge evaluation workshop, vol. 2,, 162, 167, Oct. 2013, [Last author]
English, International conference proceedings - Bursty Topics in Time Series Japanese / Chinese News Streams and their Cross-Lingual Alignment
Liyi Zheng, Takehito Utsuro, Masaharu Yoshioka
Proceedings of the 13th Conference of the Pacific Association for Computational Linguistics, Sep. 2013, [Peer-reviewed]
International conference proceedings - Query Classification by Using Named Entity Recognition Systems and Clue Keywords
Masaharu YOSHIOKA
Proceedings of the 10th NTCIR Workshop Meeting on Evaluation of Information Access Technologies: Information Retrieval, Quesiton Answering, And Cross-Lingual Information Access,, 255, 259, National Institute of Informatics (NII), Jun. 2013, [Peer-reviewed]
English - Issues for linking geographical open data of GeoNames and Wikipedia
Masaharu Yoshioka, Noriko Kando
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 7774, 375, 381, Springer, 2013, [Peer-reviewed]
English, International conference proceedings, It is now possible to use various geographical open data sources such as GeoNames and Wikipedia to construct geographic information systems. In addition, these open data sources are integrated by the concept of Linked Open Data. There have been several attempts to identify links between existing data, but few studies have focused on the quality of such links. In this paper, we introduce an automatic link discovery method for identifying the correspondences between GeoNames entries and Wikipedia pages, based on Wikipedia category information. This method finds not only appropriate links but also inconsistencies between two databases. Based on this integration results, we discuss the type of inconsistencies for making consistent Linked Open Data. © Springer-Verlag 2013. - Time Series Topic Modeling and Bursty Topic Detection of Correlated News and Twitter.
Daichi Koike, Yusuke Takahashi, Takehito Utsuro, Masaharu Yoshioka, Noriko Kando
Sixth International Joint Conference on Natural Language Processing, IJCNLP 2013, Nagoya, Japan, October 14-18, 2013, 917, 921, Asian Federation of Natural Language Processing / ACL, 2013, [Peer-reviewed] - Construction of a Japanese gazetteers for Japanese local toponym disambiguation
Masaharu Yoshioka, Takahiro Fujiwara
Proceedings of the 7th Workshop on Geographic Information Retrieval, GIR 2013, 57, 63, Association for Computing Machinery, 2013, [Peer-reviewed], [Lead author]
English, International conference proceedings, When processing toponym information in natural language text, it is crucial to have a good gazetteers. There are several well-organized gazetteers for English text, but they do not cover Japanese local toponyms. In this paper, we introduce a Japanese gazetteers based on Open Data (e.g., the Toponym database distributed by Japanese ministries, Wikipedia, and GeoNames) and propose a toponym disambiguation framework that uses the constructed gazetteers. We also evaluate our approach based on a blog corpus that contains place names with high ambiguity. - Issues for Linking Geographical Open Data of GeoNames and Wikipedia
Masaharu Yoshioka, Noriko Kando
Proceedings of the 2nd Joint International Semantic Technology Conference (JIST2012),, Dec. 2012, [Peer-reviewed], [Lead author]
English, International conference proceedings - Cross-Lingual Topic Alignment in Time Series Japanese / Chinese News
Shuo Hu, Yusuke Takahashi, Liyi Zheng, Takehito Utsuro, Masaharu Yoshioka, Noriko Kando, Tomohiro Fukuhara, Hiroshi Nakagawa, Yoji Kiyota
Proceedings of the 26th Pacific Asia Conference on Language, Information and Computation, 532, 541, Nov. 2012, [Peer-reviewed]
English, International conference proceedings - Discovery and Maintenance of Links between Wikipedia and GeoNames by Using Wikipedia Category
Masaharu Yoshioka, Yiqi Liu, Noriko Kando
情報処理学会論文誌. データベース, 5, 3, 141, 148, 一般社団法人情報処理学会, Sep. 2012, [Peer-reviewed], [Lead author]
Japanese, Scientific journal, Recently, due to the higher demand for geographic information system, it is necessary to have a good geographical database for such systems. GeoNames is one of the largest geographical database as Open Data. This database is also used for constructing web ontology by adding links to the Wikipedia page as a part of Linked Open Data. Even though, here are several attempts to find links automatically, the number of links between GeoNames and Wikipedia is not sufficient. In this paper, we propose an automatic link discovery method to use Wikipedia categories to identify the correspondence betwe... - Applying a burst model to detect bursty topics in a topic model
Yusuke Takahashi, Takehito Utsuro, Masaharu Yoshioka, Noriko Kando, Tomohiro Fukuhara, Hiroshi Nakagawa, Yoji Kiyota
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 7614, 239, 249, Springer, 2012, [Peer-reviewed]
English, International conference proceedings, This paper focuses on two types of modeling of information flow in news stream, namely, burst analysis and topic modeling. First, when one wants to detect a kind of topics that are paid much more attention than usual, it is usually necessary for him/her to carefully watch every article in news stream at every moment. In such a situation, it is well known in the field of time series analysis that Kleinberg's modeling of bursts is quite effective in detecting burst of keywords. Second, topic models such as LDA (latent Dirichlet allocation) are also quite effective in estimating distribution of topics over a document collection such as articles in news stream. However, Kleinberg's modeling of bursts is usually applied only to bursts of keywords but not to those of topics. Considering this fact, we propose how to apply Kleinberg's modeling of bursts to topics estimated by a topic model such as LDA and DTM (dynamic topic model). © 2012 Springer-Verlag Berlin Heidelberg. - Automatic information extraction of experiments from nanodevices development papers
Thaer M. Dieb, Masaharu Yoshioka, Shinjiro Hara
Proceedings of the 2012 IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012, 42, 47, 2012, [Peer-reviewed]
English, International conference proceedings, To support development processes for nanodevices, we want to utilize the information related to experiments from nanodevice development research papers. In this paper, we propose a new guideline for constructing a tagged corpus of nanodevice development papers to achieve this goal. We also propose the use of a corpus construction support tool. To speed up the construction process, we propose an approach for automatic extraction of experiment-related information from the papers. © 2012 IEEE. - Multifaceted analysis of news articles by using semantic annotated information
Masaharu Yoshioka, Noriko Kando
International Conference on Information and Knowledge Management, Proceedings, 19, 20, 2012, [Peer-reviewed]
English, International conference proceedings, We propose a novel framework that enables multifaceted analysis of news articles. This system uses semantic annotated information (e.g., person, place) as facets and can be used to construct structured queries for comparing the differences between sets of articles. - Multifaceted Analysis of News Articles by Using Semantic Annotated Information
Masaharu Yoshioka, Noriko Kando
PROCEEDINGS OF THE FIFTH WORKSHOP ON EXPLOITING SEMANTIC ANNOTATIONS IN INFORMATION RETRIEVAL, 19, +, ASSOC COMPUTING MACHINERY, 2012, [Peer-reviewed]
English, International conference proceedings, We propose a novel framework that enables multifaceted analysis of news articles. This system uses semantic annotated information (e. g., person, place) as facets and can be used to construct structured queries for comparing the differences between sets of articles. - On a Combination of Probabilistic and Boolean IR Models for Question Answering
吉岡 真治
情報処理学会論文誌, 52, 12, 3423, 3434, 情報処理学会, 15 Dec. 2011
Japanese, In order to make a good question answering (QA) system based on text database, it is better to have a good information retrieval (IR) system. To make a good IR system for QA about particular named entities, it is better to use Boolean IR model by using appropriate Boolean query. In this paper, we propose to use ABRIR (Appropriate Boolean query Reformulation for Information Retrieval) that constructs appropriate Boolean query by using list of synonyms and variation of Japanese katakana description. Evaluation results shows that ABRIR works effectively for the task of IR for QA (NTCIR-8 Geotime). - On a Combination of Probabilistic and Boolean IR Models for Question Answering
Masaharu Yoshioka
IPSJ Journal, 52, 12, 3423, 3434, 一般社団法人情報処理学会, Dec. 2011, [Peer-reviewed]
Japanese, Scientific journal, In order to make a good question answering (QA) system based on text database, it is better to have a good information retrieval (IR) system. To make a good IR system for QA about particular named entities, it is better to use Boolean IR model by using appropriate Boolean query. In this paper, we propose to use ABRIR (Appropriate Boolean query Reformulation for Information Retrieval) that constructs appropriate Boolean query by using list of synonyms and variation of Japanese katakana description. Evaluation results shows that ABRIR works effectively for the task of IR for QA (NTCIR-8 Geoti... - Evaluation of Interactive Information Access System using Concept Map.
Masaharu Yoshioka, Noriko Kando, Yohei Seki
Proceedings of the 4th International Workshop on Evaluating Information Access, EVIA 2011, National Center of Sciences, Tokyo, Japan, December 6, 2011, 20, 23, National Institute of Informatics (NII), 2011, [Peer-reviewed]
English, International conference proceedings - A Micro-analysis of Topic Variation for a Geotemporal Query.
Fredric C. Gey, Ray R. Larson, Jorge Machado, Masaharu Yoshioka
Proceedings of the 4th International Workshop on Evaluating Information Access, EVIA 2011, National Center of Sciences, Tokyo, Japan, December 6, 2011, 9, 13, National Institute of Informatics (NII), 2011, [Peer-reviewed]
English, International conference proceedings - Construction of tagged corpus for Nanodevices development papers
Thaer M. Dieb, Masaharu Yoshioka, Shinjiro Hara
Proceedings - 2011 IEEE International Conference on Granular Computing, GrC 2011, 167, 170, IEEE, 2011, [Peer-reviewed]
English, International conference proceedings, Nanodevices development process is not well systematized. In order to support this development process, we have been working on an experimental record management system that aims at analyzing previous manufacturing experiments to provide an insight on future manufacturing. However, we found that experimental records do not have enough information for detailed analysis. In this paper, we propose an approach to extract extra information (metadata) from Nanodevices development papers that can enhance the analysis of experimental results. © 2011 IEEE. - ABRIR at NTCIR-9 GeoTime Task Usage of Wikipedia and GeoNames for Handling Named Entity Information.
Masaharu Yoshioka
Proceedings of the 9th NTCIR Workshop Meeting on Evaluation of Information Access Technologies: Information Retrieval, Question Answering and Cross-Lingual Information Access, NTCIR-9, National Center of Sciences, Tokyo, Japan, December 6-9, 2011, National Institute of Informatics (NII), 2011, [Peer-reviewed] - NTCIR9-GeoTime Overview - Evaluating Geographic and Temporal Search: Round 2.
Fredric C. Gey, Ray R. Larson, Jorge Machado, Masaharu Yoshioka
Proceedings of the 9th NTCIR Workshop Meeting on Evaluation of Information Access Technologies: Information Retrieval, Question Answering and Cross-Lingual Information Access, NTCIR-9, National Center of Sciences, Tokyo, Japan, December 6-9, 2011, National Institute of Informatics (NII), 2011, [Peer-reviewed] - Knowledge exploratory project for nanodevice design and manufacturing
Masaharu Yoshioka, Katsuhiro Tomioka, Shinjiroh Hara, Takashi Fukui
iiWAS2010 - 12th International Conference on Information Integration and Web-Based Applications and Services, 871, 874, ACM, 2010, [Peer-reviewed]
English, International conference proceedings, We are developing a framework for knowledge creation in nanodevice development, based on collaboration between nanodevice engineers and computer science researchers. Development of nanodevices requires a variety of knowledge
some of this knowledge is tacit, based on the user's experience. Therefore, it is difficult to become a good engineer in this development process. We propose the concept of "Evidence-based experiment planning" and develop a process for supporting experiment planning in nanodevice development. This system applies knowledge discovery techniques to records of previous experiments to extract experienced engineers' tacit knowledge. Copyright 2010 ACM. - Extracting Concerns and Reports on Crimes in Blogs
Yusuke Abe, Takehito Utsuro, Yasuhide Kawada, Tomohiro Fukuhara, Noriko Kando, Masaharu Yoshioka, Hiroshi Nakagawa, Yoji Kiyota, Masatoshi Tsuchiya
ACTIVE MEDIA TECHNOLOGY, 6335, 498, +, SPRINGER-VERLAG BERLIN, 2010, [Peer-reviewed]
English, International conference proceedings, Among other domains and topics on which some issues are frequently argued in the blogosphere, the domain of crime is one of the most seriously discussed by various kinds of bloggers. Such information on crimes in blogs is especially valuable for those who are not familiar with tips for preventing being victimized. This paper proposes a framework of extracting people's concerns and reports on crimes in their own blogs. In this framework, we solve two tasks. In the first task, we focus on experts in crime domain and address the issue of extracting concerns on crimes such as tips for preventing being victimized. In the evaluation of this first task, we show that we successfully rank expert bloggers high in the results of blog feed retrieval. In the second task, on the other hand, we focus on victims of criminal acts, and address the issue of extracting reports on being victimized In the evaluation of this second task, we show that we outperform blog post ranking based on the search engine API by incorporating dependency relations for identifying victims' blog posts. - On a Combination of Probabilistic and Boolean IR Models for Question Answering
Masaharu Yoshioka
INFORMATION RETRIEVAL TECHNOLOGY, 6458, 588, 598, SPRINGER-VERLAG BERLIN, 2010, [Peer-reviewed]
English, International conference proceedings, To make a good question answering (QA) system based on a text database, it is preferable to have a good information retrieval (IR) system that can find appropriately relevant document sets for a given query. To make a good IR system for QA about particular named entities (NEs), it is preferable to use a Boolean IR model that uses appropriate Boolean queries with the NE information. In this paper, we propose to use appropriate Boolean query reformulation for information retrieval (ABRIR) for this problem. In this system, an appropriate list of synonyms and variations of Japanese katakana descriptions of a given query are used to construct the Boolean query. Evaluation results show that ABRIR works effectively for the IR. task in QA. - Cross-lingual analysis of concerns and reports on crimes in blogs
Hiroyuki Nakasaki, Yusuke Abe, Takehito Utsuro, Yasuhide Kawada, Tomohiro Fukuhara, Noriko Kando, Masaharu Yoshioka, Hiroshi Nakagawa, Yoji Kiyota
Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, 40, 315, 320, Springer, 2010, [Peer-reviewed]
English, International conference proceedings, Among other domains and topics on which some issues are frequently argued in the blogosphere, the domain of crime is one of the most seriously discussed by various kinds of bloggers. Such information on crimes in blogs is especially valuable for outsiders from abroad who are not familiar with cultures and crimes in foreign countries. This paper proposes a framework of cross-lingually analyzing people's concerns, reports, and experiences on crimes in their own blogs. In the retrieval of blog feeds/posts, we take two approaches, focusing on various types of bloggers such as experts in the crime domain and victims of criminal acts. © Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering 2010. - On a Combination of Probabilistic and Boolean IR Models for GeoTime Task.
Masaharu Yoshioka
Proceedings of the 8th NTCIR Workshop Meeting on Evaluation of Information Access Technologies: Information Retrieval, Question Answering and Cross-Lingual Information Access, NTCIR-8, National Center of Sciences, Tokyo, Japan, June 15-18, 2010, 154, 158, National Institute of Informatics (NII), 2010, [Peer-reviewed] - NSContrast: An Exploratory News Article Analysis System that Characterizes the Differences between News Sites
Masaharu Yoshioka
SIGIR2009 Workshop on Information Access in a Multilingual World, 25, 29, Jul. 2009, [Peer-reviewed]
English, International conference proceedings - NSContrast: Characterizing the Difference of Interest among Multiple News Sites
Masaharu Yoshioka
INTERACT 2009 Workshop on Culture and Technologies for Social Interaction, http://sites.google.com/site/crosscultureworkshopinteract09/position-papers-1/10-Yoshioka.pdf-29, 2009, [Peer-reviewed], [Lead author]
English, International conference proceedings - Web content summarization using social bookmarks: a new approach for social summarization
K. Takahashi, A. Sugiyama, Y. Shimomura, T. Tateyama, R. Chiba, M. Yoshioka, H. Takeda
WIDM '08: Proceeding of the 10th ACM workshop on Web information and data management, 103, 110, Oct. 2008, [Peer-reviewed]
English, International conference proceedings - Creating Vital Information Technologies for the Info-plosion Era : Beyond Keyword Search - Info-plosion Search for Mining Valuable Unknowns -
TORISAWA Kentaro, NAKAGAWA Hiroshi, KUROHASHI Sadao, INUI Kentaro, YOSHIOKA Masaharu, FUJII Atsushi, KITSUREGAWA Masaru
IPSJ Magazine, 49, 8, 890, 896, Information Processing Society of Japan (IPSJ), 15 Aug. 2008
Japanese - Web-based knowledge database construction method for supporting design
Kiyotaka Takahashi, Aki Sugiyama, Yoshiki Shimomura, Takeshi Tateyama, Ryosuke Chiba, Masaharu Yoshioka, Hideaki Takeda
Proceedings of the 10th International Conference on Information Integration and Web-based Applications and Services, iiWAS 2008, 18th, 575, 578, 2008, [Peer-reviewed]
English, International conference proceedings, Recently, designers have been required solving comprehensive problem becoming greater and more complicated. In relation to this background, we have proposed Universal Abduction Studio (UAS)
a computer environment that synthetically supports creative design. However, it is difficult for a designer to manually acquire multiple domain knowledge required for UAS. Therefore, we propose a Web-based knowledge database construction method for supporting design by UAS in this paper. © 2008 ACM. - IR Interface for Contrasting Multiple News Sites
Masaharu Yoshioka
Proceedings of the Asia Information Retrieval Symposium 2008, -, -, 516, 521, Jan. 2008, [Peer-reviewed]
English, International conference proceedings - Analyzing Multiple News Sites by Contrasting Articles
Masaharu Yoshioka
SITIS 2008: 4TH INTERNATIONAL CONFERENCE ON SIGNAL IMAGE TECHNOLOGY AND INTERNET BASED SYSTEMS, PROCEEDINGS, 45, 51, IEEE COMPUTER SOC, 2008, [Peer-reviewed]
English, International conference proceedings, Today, there is access to large numbers of news sites in different countries, and there are several experimental systems, such as Newsblaster and NewsExplorer that integrate news articles about a particular event from multiple news sites. These systems enable a good understanding of particular events by using multiple news sites, but they ignore the characteristics of the various news sites. To characterize the differences between news sites, the News Site Contrast (NSContrast) system has been proposed. This system analyzes multiple news sites using the concept of contrast-set mining. However NSContrast has only limited analysis functions and is not mature enough for evaluation via user experiments. Therefore, in this paper, we analyze contrast set mining results for NSContrast, aiming to understand the requirements for extracting useful information that also reflects the interests of different countries. Based on this analysis, a new NSContrast is introduced and applied to a news article database constructed from multiple news sites in different countries for user experimentation. - Web-Based Knowledge Database Construction Method for Supporting Design
Kiyotaka Takahashi, Aki Sugiyama, Yoshiki Shimomura, Takeshi Tateyama, Ryosuke Chiba, Masaharu Yoshioka, Hideaki Takeda
PRACTICAL ASPECTS OF KNOWLEDGE MANAGEMENT, PROCEEDINGS, 5345, 173, +, SPRINGER-VERLAG BERLIN, 2008, [Peer-reviewed]
English, International conference proceedings, In recent years, comprehensive problem solving by artefact designers has been required as demands related to artefacts become greater and more complicated. In relation to this background, we have proposed Universal Abduction Studio (UAS), a computer environment that synthetically supports a creative design. However, in general, it is difficult for a designer to acquire multiple domain knowledge because this knowledge is accumulated manually. This paper proposes a Web-based knowledge database construction method for supporting design using UAS. - 人間の知的活動支援へのアプローチ特集[インテリジェンス系]マイナーなテキスト情報をも活用する情報アクセス技術
吉岡真治
ヒューマンインタフェース学会誌, 9, 4, 269-274, Nov. 2007, [Invited]
Japanese, Scientific journal - Support for Creative Design by Abduction
TAKEDA Hideaki, SHIMOMURA Yoshiki, YOSHIOKA Masaharu
Journal of Japan Society for Design Engineering, 42, 10, 558, 564, 日本設計工学会, 05 Oct. 2007
Japanese - Tagging for intelligent processing of design information
Hideaki Takeda, Yutaka Fujimoto, Masaharu Yoshioka, Yoshiki Shimomuraz, Kengo Morimoto, Wataru Oniki
NEW FRONTIERS IN ARTIFICIAL INTELLIGENCE, 3609, -, 216, +, SPRINGER-VERLAG BERLIN, 2007, [Peer-reviewed]
English, International conference proceedings, This paper describes how to add tags to design documents in order to extract knowledge from information for intelligent design support. Our project called Universal Abduction Studio (UAS) aims to build a new design support system that supports conceptual design by dynamically integrating knowledge in different design domains. This paper focuses on knowledge description form which can be used to capture knowledge from text-based information and then be used for inference for creative design. We propose so-called design knowledge document containing both human-readable texts and machine-readable knowledge such as propositions and rules. - Evaluating topic difficulties from the viewpoint of query term expansion
Masaharu Yoshioka
INFORMATION RETRIEVAL TECHNOLOGY, PROCEEDINGS, 4182, 390, 403, SPRINGER-VERLAG BERLIN, 2006, [Peer-reviewed]
English, Scientific journal, Query term expansion is an important technique for achieving higher retrieval performance. However, since many factors affects the quality of this technique, it is difficult to evaluate this technique in isolation. Feature quantities that characterize the quality of the initial query are defined in this study for evaluating topic difficulties from the viewpoint of query term expansion. I also briefly review the result of the NTCIR-5 query term expansion subtask that uses these quantities for evaluating the effectiveness of the query term expansion techniques. I also describe detailed analysis results on the effect of query term expansion based on topic-by-topic analysis. - Fundamental Plot for a Designer Support Environment based on Abduction
SHIMOMURA Yoshiki, YOSHIOKA Masaharu, TAKEDA Hideaki, TOMIYAMA Tetsuo
TRANSACTIONS OF THE JAPAN SOCIETY OF MECHANICAL ENGINEERS Series C, 72, 713, 274, 281, The Japan Society of Mechanical Engineers, 2006
Japanese, This paper describes a new design support system that supports conceptual or creative design by dynamically integrating knowledge in different design domains. We argue that abduction for integrating theories can be a basic principle to formalize such design processes. Based on this principle, we propose Universal Abduction Studio, a design environment in which designers combine different theories to arrive at better design. In this new approach to computational support of conceptual design, the system should offer various types of abductive reasoning from which designers can select an interesting design method. We also discuss technologies to implement UAS and in this paper we propose to use analogical reasoning as abductive reasoning to discover relationships between knowledge from different sources. We demonstrate that a prototype system can discover a new idea in a design example taken from a real design activity. - Research on Information Retrieval System that Supports Keyword Selection based on Generalized Concept and Coverage
YOSHIOKA Masaharu, HARAGUCHI Makoto
Transactions of the Japanese Society for Artificial Intelligence, 20, 4, 270, 280, The Japanese Society for Artificial Intelligence, 01 Nov. 2005
Japanese, It is not easy for a user of an Information Retrieval (IR) system to select an appropriate keyword set to represent his or her specific information need. Therefore, many IR systems can modify keyword sets by estimating the user's particular requirement. Even though such IR systems have better retrieval performance, the complicated estimation process entailed by a large number of keywords makes it difficult for a user to understand how the system behaves. Therefore, we used a thesaurus for query expansion. To select an appropriate keyword set, we proposed two concepts: ``adaptive generalization'', which estimates an appropriate generalization level of the given keywords by using relevant document information, and ``purpose-oriented concept structure modification'', which selects relevant keywords from a predefined synonym set in a thesaurus. Because query expansion based on a thesaurus aims to find new keywords that are complementary to the initial keywords, we proposed to use this method to construct a Boolean query formula to represent the user's information need. We proposed a new IR system called ``appropriate Boolean query reformulation for IR with adaptive generalization'' (ABRIR-AG) to support Boolean query formation. In ABRIR-AG, we reformulated a user-given Boolean query by using a small number of relevant documents. Finally, to evaluate its effectiveness, we evaluated ABRIR-AG by using a large-scale test collection containing WWW documents. - An Architecture for Designers' Support Systems with Knowledge-embedded Documents
Hideaki Takeda, Masaharu Yoshioka, Yoshiki Shimomura, Yutaka Fujimoto, Kengo Morimoto, Wataru Oniki
Proceedings of the 15th International Conference on Engineering Design -ICED05- (CD-ROM), 2005, [Peer-reviewed] - An appropriate Boolean query reformulation interface for information retrieval based on adaptive generalization
M Yoshioka, M Haraguchi
International Workshop on Challenges in Web Information Retrieval and Integration, Proceedings, 145, 150, IEEE COMPUTER SOC, 2005, [Peer-reviewed]
English, International conference proceedings, When implementing an IR system that can support comprehensive searches from a wide variety of documents, it is crucial to have a mechanism for selecting appropriate query formulations. For this purpose, many IR systems can modify query terms by estimating the user's information need. However because modified query terms are usually represented in a complicated form, it is difficult to judge how appropriate they are. In this research, we assume all relevant documents should contain words that correspond to concepts that the user would like to retrieve. Therefore, a Boolean query formula is one good way to represent a user's information need with high readability. To find the required concepts in a relevant document set, we have previously proposed a new IR system that calculates appropriate generalized concepts to represent the user's need. In this study, we implement a user interface that supports reformulation of IR queries by using abstract concepts. - On a combination of probabilistic and Boolean IR models for WWW document retrieval
Masaharu Yoshioka, Makoto Haraguchi
ACM Transactions on Asian Language Information Processing, 4, 3, 340, 356, 2005, [Peer-reviewed]
English, International conference proceedings, Even though a Boolean query can express the information need precisely enough to select relevant documents, it is not easy to construct an appropriate Boolean query that covers all relevant documents. To utilize a Boolean query effectively, a mechanism to retrieve as many as possible relevant documents is therefore required. In accordance with this requirement, we propose a method for modifying a given Boolean query by using information from a relevant document set. The retrieval results, however, may deteriorate if some important query terms are removed by this reformulation. A further mechanism is thus required in order to use other query terms that are useful for finding more relevant documents, but are not strictly required in relevant documents. To meet this requirement, we propose a new method that combines the probabilistic IR and the Boolean IR models. We also introduce a new IR system - called appropriate Boolean query reformulation for information retrieval (ABRIR) - based on these two methods and the Okapi system. ABRIR uses both a word index and a phrase index formed from combinations of two adjacent noun words. The effectiveness of these two methods was confirmed according to the NTCIR-4 Web test collection. © 2005 ACM. - Towards construction of evaluation framework for query expansion
M Yoshioka
INFORMATION RETRIEVAL TECHNOLOGY, PROCEEDINGS, 3689, 647, 652, SPRINGER-VERLAG BERLIN, 2005, [Peer-reviewed]
English, Scientific journal, Query expansion is an important technique for achieving higher retrieval performance. However, a good framework for evaluating this technique in isolation does not, exist because the effect of query expansion depends on the quality of the initial query. Feature quantities that characterize the quality of the initial query are defined in this study. Correlations between these quantities and retrieval performance were analyzed using the NTCIR-4 web test collection. - Physical concept ontology for the knowledge intensive engineering framework
M Yoshioka, Y Umeda, H Takeda, Y Shimomura, Y Nomaguchi, T Tomiyama
ADVANCED ENGINEERING INFORMATICS, 18, 2, 95, 113, ELSEVIER SCI LTD, Apr. 2004, [Peer-reviewed]
English, Scientific journal, Knowledge intensive engineering aims at flexible applications of a variety of product life cycle knowledge. much as design, manufacturing. operations, maintenance, and recycling. Many engineering domain theories are organized and embedded within CAD and CAE tools and engineering activities can be formalized as modeling operations to them. Since most of domain theories deal with the physical world and can be associated with physical concepts. a physical concept ontology can form a common ontology to integrate engineering models that are formed based on domain theories. This paper reports a physical ontology-based support system for knowledge intensive engineering called Knowledge intensive Engineering Framework (KIEF) to integrate multiple engineering models and to allow more flexible use of them. First the paper describes the physical ontology as the core of KIEF and an ontology-based reasoning system called a pluggable metamodel mechanism, to integrate and maintain relationships among these models. The pluggable metamodel mechanism uses a metamodel that represents the designer's mental model about a design object as a concept network model. The designer builds and decomposes a functional hierarchy from functional specifications with an FBS (Function-Behavior-State) modeler. He/She then maps the functional hierarchy into a metamodel using physical features that are building blocks for conceptual design. Then, the pluggable metamodel mechanism enriches the information contained in the metamodel by using causal dependency knowledge about the physical world and by budding and analzing various engineering models. We demonstrate the power of KIEF by illustrating a design case performed on KIEF. (C) 2004 Published by Elsevier Ltd. - A methodology for IR query clarification based on adaptive generalization
M Yoshioka, M Haraguchi
8TH WORLD MULTI-CONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL I, PROCEEDINGS, I, 226, 230, INT INST INFORMATICS & SYSTEMICS, 2004
English, International conference proceedings, We can use various types of information retrieval (IR) system to access large quantities of text data. These systems usually request users to input keywords to represent their retrieval intention. Since, it is difficult for a user to select appropriate keywords, many IR systems have the capability of modifying keywords by estimating the user's intention. However because modified keywords are usually represented in a complicated form, it is difficult to judge how appropriate they are. In this research, we assume all relevant documents should have words that correspond to concepts that the user would like to retrieve. Therefore, a boolean query formula is one good way to represent user's intent with higher readability. In order to find out these concepts in a relevant document set, we have already proposed a new IR system that calculates appropriate generalized concept to represent user's intention. In this study, we implement a user interface that supports to clarify IR query by using abstract concepts. - Towards constructing story databases using maximal analogies between stories
M Yoshioka, M Haraguchi, A Mizoe
INTUITIVE HUMAN INTERFACES FOR ORGANIZING AND ACCESSING INTELLECTUAL ASSETS, 3359, 243, 255, SPRINGER-VERLAG BERLIN, 2004, [Peer-reviewed]
English, Scientific journal, In order to construct story databases, it is crucial to have an effective index that represents the plot and event sequences in a document. For this purpose, we have already proposed a method using the concept of maximal analogy to represent a generalized event sequence of documents with a maximal set of events. However, it is expensive to calculate a maximal analogy from documents with a large number of sentences. Therefore, in this paper, we propose an efficient algorithm to generate a maximal analogy, based on graph theory, and we confirm its effectiveness experimentally. We also discuss how to use a maximal analogy as an index for a story database, and outline our future plans. - Universal abduction studio -- proposal of a design support environment for creative thinking in design --
Hideaki Takeda, Hiromitsu Sakai, Yutaka Nomaguchi, Masaharu Yoshioka, Yoshiki Shimomura, Tetsuo Tomiyama
A. Folkman, K. Gralen, M. Norell, and U. Sellgren, editors, The Fourteenth International Conference on Engineering Design (ICED 03), Stockholm, August 2003, Aug. 2003, [Peer-reviewed]
English - Construction of Personalized and Purpose-Oriented Thesaurus
Masaharu Yoshioka, Makoto Haraguchi
Proc. of Asian Association for Lexicography' 03 (ASIALEX), 461, 466, 2003 - Towards an organization and access method of story databases
M Haraguchi, M Yotsutani, M Yoshioka
7TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL V, PROCEEDINGS, V, 213, 216, INT INST INFORMATICS & SYSTEMICS, 2003
English, International conference proceedings, To classify a document set from various viewpoints, we extract a possible abstract document from given similar documents. The abstract one is a common generalization of the original documents. Although we can present an efficient algorithm to find it under some pruning rule, it is generally very hard to perform such a task of extracting abstract documents. It is therefore necessary to reduce the numbers of events appearing in documents. For this purpose, we regard one document as a network of events, present a pseudo-PageRank model for calculating the importance of events, analyze and compare it with a standard text summarization method, and finally reports the present approach to improve the pseudo-PageRank. - Knowledge Management System for Problem Solving: Integration of Document Information and Formalized Knowledge
Masaharu Yoshioka, Yasuhiro Shamoto
Volume 1: 23rd Computers and Information in Engineering Conference, Parts A and B, ASMEDC, 01 Jan. 2003, [Peer-reviewed], [Lead author]
International conference proceedings, In engineering activities various types of problems have to be solved. Most of these problems are routine problems and some are complicated problems for which it is difficult to formalize all the problem-solving processes. We propose a knowledge management system for supporting these problem solving processes. One example is using information applicable to new and complicated problems that has resulted from solving similar problems in the past. In addition, in such complicated problem solving processes, most of the decomposed sub-problems can be handled as routine problems. In this paper, we propose use of formalized knowledge and document information that represents the history of problem-solving processes to support the solution of both types of problem. For that purpose, we propose the Documentation Management system for Problem Solving (DMaPS), which supports a user solving a problem by document processing. In this support process, the system retrieves related documents for the user’s problems using formalized knowledge and information retrieval techniques. After retrieving related documents, the user modifies them to fit their problem. The modified information is associated with the formalized knowledge and is stored as documents for future reuse. - Abduction for Creative Design
Tetsuo Tomiyama, Hideaki Takeda, Masaharu Yoshioka, Yoshiki Shimomura
Computational Synthesis: From Basic Building Blocks to High Level Functionality Papers from the 2003 AAAI Symposium, 239, 246, 2003
International conference proceedings - Abduction for creative design
Tetsuo Tomiyama, Hideaki Takeda, Masaharu Yoshioka, Yoshiki Shimomura
Proceedings of the ASME Design Engineering Technical Conference, 3, 543, 552, 2003, [Peer-reviewed]
International conference proceedings, While abduction is considered crucial for design in general, this paper focuses on the role of abduction to integrate knowledge assuming that creative design can come from innovative combination of existing knowledge. Based on Schurz's classification of abductive reasoning, the paper identifies that abduction for integrating theories can be performed by a special type of abduction called second order existential abduction. The paper then analyzes refrigerator design cases to understand how knowledge is used and shows that abduction is indeed central to design. It also discusses that knowledge structure is a key concept in abduction for integration. - Development of an information retrieval system suitable for large-scale scholarly databases
Oyama, K., Kageura, K., K, o, N., Kimura, M., Maruyama, K., Yoshioka, M., Takahashi, K.
Systems and Computers in Japan, 34, 6, 44, 56, Jan. 2003, [Peer-reviewed]
English, Scientific journal - Analysis of Japanese Deverbal Compounds Based on Lexical Conceptual Structure
TAKEUCHI Koichi, UCHIYAMA Kiyoko, YOSHIOKA Masaharu, KAGEURA Kyo, KOYAMA Teruo
Transactions of Information Processing Society of Japan, 43, 5, 1446, 1456, Information Processing Society of Japan (IPSJ), 15 May 2002
Japanese, In this paper, we describe a principled approach for analyzing relations between constituent words of compound nouns, specifically those whose heads are deverbal nouns, based on the classification of deverbal nouns by their lexical conceptual structure (LCS) and the classification of nouns in modifier position vis-a-vis LGS of head deverbal nouns. There are two kinds of relations of compounds with head deverbal nouns. The one is that a modifier noun becomes an argument of deverbal head and the other is that a modifier becomes an adjunct. It is an important starting point for analyzing relations to disambiguate the two kinds of relations. Through the qualitative analysis of the data and the experimental evaluation of 1231 compound nouns, we show that the use of LCS as the theoretical basis is very promising for constructing compound analyzer. - NTCIR from Participants' Viewpoints(
NTCIR from the View Point of Participant)
OGAWA Yasushi, SASAKI Yutaka, MASUYAMA Shigeru, MURATA Masaki, YOSHIOKA Masaharu, Yasushi Ogawa, Yutaka Sasaki, Shigeru Masuyama, Masaki Murata, Masaharu Yoshioka
Journal of Japanese Society for Artificial Intelligence, 17, 3, 306, 311, The Japanese Society for Artificial Intelligence, 01 May 2002
Japanese - 人工物工学研究の展開 第18報 -Universal Abduction Studioの提案-
坂井宏充, 野間口大, 吉岡真治, 武田英明, 下村芳樹, 冨山哲男
日本機械学会第12回設計工学・システム部門講演会講演論文集, Apr. 2002
Japanese, Research society - Document-based Design Process Knowledge Management for Knowledge Intensive Engineering
Y. Nomaguchi, M. Yoshioka, T. Tomiyama
From Knowledge Intensive CAD to Knowledge Intensive Engineering, Kluwer Academic Publishers, Apr. 2002, [Peer-reviewed]
English, Scientific journal - Construction of a Large Scale Test Collection NTCIR-2 : The Effect of Additional Interactive Search and Cross-Lingual Pooling
KURIYAMA KAZUKO, YOSHIOKA MASAHARU, KANDO NORIKO
情報処理学会論文誌データベース(TOD), 43, 2, 48, 59, Information Processing Society of Japan (IPSJ), 15 Mar. 2002
Japanese, The purposes of this study are to examine whether there is an effect on the relative evaluation of the IR systems using the relevance judgments of the test collection NTCIR-2 made by the pooling method and additional interactive searches, and to investigate whether the NTCIR-2 is effective for evaluating the IR systems, the search results of which were not used for the pooling. We carried out experiments using different lists of relevance judgments and search results submitted for the test of the 2nd NTCIR Workshop. First, we evaluated the search results using the list of the final relevance judgments F of NTCIR-2 and F - I, that is, the F without the unique relevant documents found by the additional interactive searches I. Second, we made pools from the search results for each of the sub-tasks and evaluated the search results using the pools as lists of relevance judgments. Third, we evaluated the search results using F - S, that is, the F without the unique relevant documents found by an IR system S in order to simulate evaluation of the IR system which was not used for the pooling. Almost the same rankings of the search results were produced by using the pools as lists of relevance judgments for system evaluation. When the search results by an JR system were not used for the pooling, there is a very little effect on evaluation of the system itself and relative evaluation among it and other systems. Therefore our results verified the reliability of test collection as an evaluation tool, which was based on pooling method. - Categorising deverbal nouns based on lexical conceptual structure for analysing Japanese compounds
K Takeuchi, K Uchiyama, M Yoshioka, K Kageura, T Koyama
2001 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS, VOLS 1-5, 904, 909, IEEE, 2002, [Peer-reviewed]
English, International conference proceedings, This paper proposes a method to describe conceptual meanings of Japanese deverbal nouns for the purpose of analysing Japanese compound nouns. A lexical meaning of each constituent word is very important since it deeply influences a word formation of compound words. Therefore we intend to use the lexical meanings of constituent words to build a Japanese compound noun analyser. As a first step to describe lexical meanings of words, we have established a simple semantic structure based on lexical conceptual structure (we refer to the structure we establish by TLCS1 in this paper). After the analysing 250 deverbal nouns that appeared in technical terms, we found that the only 14 types of TLCSes cover various deverbal nouns. We will explain in this paper the types of TLCS and also show the procedure of assigning TLCSes to deverbals. In order to evaluate the efficacy of TLCS, we made use of the accuracy of simple compound noun analyser for analysing technical terms. The result was very good, which shows that our proposed method is very promising. - A proposal of CAD mechanism for design knowledge management
Yutaka Nomaguchi, Masaharu Yoshioka, Tetsuo Tomiyama
Transactions of the Japanese Society for Artificial Intelligence, 17, 1, 104, 113, 2002
Scientific journal, In this paper, we propose a fundamental idea of a new CAD mechanism to facilitate design knowledge management. This mechanism encourages a designer to externalise his/her knowledge during a design process and facilitates sharing and reuse of such externalised design knowledge in later stages. We also describe the implementation of this idea called DDMS (Design Documentation Management System). DDMS works as a front end to KIEF (Knowledge Intensive Engineering Framework), which we have been developing. We also illustrate an example of machining tool design to demonstrate the features of DDMS. - Discovery of maximal analogies between stories
M Haraguchi, S Nakano, M Yoshioka
DISCOVERY SCIENCE, PROCEEDINGS, 2534, 324, 331, SPRINGER-VERLAG BERLIN, 2002
English, Scientific journal, Given two documents in the form of texts, we present a notion of maximal analogy representing a generalized event sequence of documents with a maximal set of events. They are intended to be used as extended indices of documents to automatically organize a document database from various viewpoints. The maximal analogy is defined so as to satisfy a certain consistency condition and a cost condition. Under the consistency condition, a term in an event sequence is generalized to more abstract term independently of its occurrence positions. The cost condition is introduced so that meaningless similarities between documents are never concluded. As the cost function is monotone, we can present an optimized bottom-up search procedure to discover a maximal analogy under an upper bound of cost. We also show some experimental results based on which we discuss a future plan. - A Knowledge Operation model of Synthesis
TOMIYAMA Tetsuo, YOSHIOKA Masaharu, TSUMAYA Akira
Engineering Design Synthesis - Understanding, Approaches, and Tools, Springer-Verlag, pp.67-90, Jan. 2002
English, Scientific journal - 設計知識管理のためのCADアーキテクチャーの提案
野間口大, 吉岡真治, 冨山哲男
人工知能学会論文誌, 17, 104, 113, 社団法人 人工知能学会, 2002, [Peer-reviewed]
Scientific journal, In this paper, we propose a fundamental idea of a new CAD mechanism to facilitate design knowledge management. This mechanism encourages a designer to externalise his/her knowledge during a design process and facilitates sharing and reuse of such externalised design knowledge in later stages. We also describe the implementation of this idea called DDMS (Design Documentation Management System). DDMS works as a front end to KIEF (Knowledge Intensive Engineering Framework), which we have been developing. We also illustrate an example of machining tool design to demonstrate the features of DDMS. - Proposal of an Integrated Design Support Environment Based on the Model of Synthesis
Masaharu Yoshioka, Yutaka Nomaguchi, Tetsuo Tomiyama
Volume 2B: 27th Design Automation Conference, American Society of Mechanical Engineers, 09 Sep. 2001, [Peer-reviewed], [Lead author]
International conference proceedings,Abstract
We propose a new integrated computational environment, called a Knowledge Intensive Engineering Framework (KIEF), to support design object modeling, during engineering design process that requires a variety of design object models, such as a geometric model, a control model, and a finite element model. KIEF is capable of guiding designers with design process knowledge based on a model of synthesis, of integrating multiple design object models, and of maintaining consistency among these models. The design process knowledge is a result of a research project we conduct to establish a model of synthesis. This model formalizes design process knowledge in a design process as operations to a multiple model-based reasoning system, such as KIEF, and from a language of synthesis. We describe a prototype system that incorporates design process knowledge to KIEF and illustrate an example design on it. - A Theory of Design Object Knowledge Operations
YOSHIOKA Masaharu, TOMIYAMA Tetsuo
Journal of the Japan Society for Precision Engineering, 67, 9, 1490, 1496, The Japan Society for Precision Engineering, 05 Sep. 2001
Japanese, In a design process, a designer uses various kinds of design object knowledge, such as physical laws, mechanism library, and manufacturing knowledge. Therefore, it is necessary for a CAD system to handle large scale knowledge bases to support this process. However, it is difficult to use a large scale knowledge base without knowledge about operations of the knowledge itself. This paper proposes a theory of design knowledge operations. First, we analyze the results of design experiments that are records of real design processes. Second, we formalize design knowledge operations based on General Design Theory. Finally we discuss the generality of the proposed formalism of design knowledge operations. - A Model of Synthesis: A System of Multiple Theories
TOMIYAMA Tetsuo, HEW Ken P, MURAKAMI Tamotsu, YOSHIOKA Masaharu, TAKEDA Hideaki, WASHIO Takeshi, TSUMAYA Akira, UMEDA Yasushi
Proceedings of the International CIRP Design Seminar 2001, pp.1-6, Jun. 2001, [Peer-reviewed]
English, International conference proceedings - Development of an Information Retrieval System Suitable for Large-Scale Scholarly Databases
OYAMA Keizo, KAGEURA Kyo, KANDO Noriko, KIMURA Masaru, MARUYAMA Katsumi, YOSHIOKA Masaharu, TAKAHASHI Kazumichi
The Transactions of the Institute of Electronics,Information and Communication Engineers., 84, 6, 658, 670, The Institute of Electronics, Information and Communication Engineers, 01 Jun. 2001
Japanese, インターネットの発展に伴い, WWWを介した本格的な情報検索サービスの堤供が求められているが, 既存のWWW情報検索システムでは, 検索機能やデータベースの構造・規模などに制約が多い.本論文の情報検索システムは実際の情報検索サービスで運用されているものであり, 大規模で複雑な構造をもつテキスト型の学術情報データベースに対し, 高度な検索機能を, ラインモードとWWWインタフェースにより提供している.本論文では, まず情報検索機能の概要と設計について, 特に集合演算, シソーラス検索, 階層構造検索, 複数データベースの統合検索などの特徴的な機能について述べている.次に, その実現のための実装技術として, 検索エンジンの特長, セッション管理方式, 大規模データベースの扱い, サーバ及びプロセスの構成などを示し, 現状のシステムについての運用上の評価を行い, 最後に今後の課題と関連研究について議論している. - 設計シミュレータを用いた設計文書作成支援
野間口大, 冨山哲男, 吉岡真治
2001年度精密工学会春季大会学術講演会講演論文集, Apr. 2001
Japanese, Research society - A general framework for modelling of synthesis integration of theories of synthesis
H Takeda, M Yoshioka, T Tomiyama
DESIGN RESEARCH - THEORIES, METHODOLOGIES, AND PRODUCT MODELLING, 307, 314, PROFESSIONAL ENGINEERING PUBLISHING LTD, 2001
English, International conference proceedings - Design for upgradable products considering future uncertainty
Y Umemori, S Kondoh, Y Umeda, Y Shimomura, M Yoshioka
SECOND INTERNATIONAL SYMPOSIUM ON ENVIRONMENTALLY CONSCIOUS DESIGN AND INVERSE MANUFACTURING, PROCEEDINGS, 87, 92, IEEE COMPUTER SOC, 2001
English, International conference proceedings, Conventional products tend to be thrown away because of functional obsoleteness before reaching the end of physical life of the products, This is one of the main causes of the mass disposal problem. In order to solve this problem, this paper proposes a design methodology for upgradable products, Especially, this paper discusses a method dealing with uncertainty caused by long-term planning in the design methodology for upgradability Namely, the proposed method deals with uncertainty as ranges of parameter values, and derives a design solution that realizes required upgrades, while adapting to estimated uncertainty. This method makes the products robust and tolerant against the uncertainty and, therefore, makes design for upgradability more feasible. - Verification of a model of synthesis - The methods for verification and results
A Tsumaya, Y Nomaguchi, M Yoshioka, H Takeda, T Murakami, T Tomiyama
DESIGN RESEARCH - THEORIES, METHODOLOGIES, AND PRODUCT MODELLING, 229, 236, PROFESSIONAL ENGINEERING PUBLISHING LTD, 2001
English, International conference proceedings - A Model of Synthesis from the Viewpoint of Design Knowledge Operations
TOMIYAMA Tetsuo, YOSHIOKA Masaharu, TSUMAYA Akira
Proceedings of the 1st International Conference on Mechanocal Engineering (ICME2000), CD-ROM (6 pages), Nov. 2000, [Peer-reviewed]
English, International conference proceedings - IR/IE/Summarisation Evaluation Projects in Japan
Kageura,K, Fukushima, T, Kando, N, Okumura, M, Sekine, S, Kuriyama, K, Takeuchi, K, Yoshioka, M, Koyama, T, Isahara, H
Proceedings of LREC 2000 Workshop on Using Evaluation within HLT Programs, Athens, Greece, May, 2000, 19, 22, May 2000, [Peer-reviewed]
English, International conference proceedings - Recent advances in automatic term recognition: Experiences from the NTCIR workshop on information retrieval and term recognition
Kyo Kageura, Masaharu Yoshioka, Koichi Takeuchi, Teruo Koyama, Keita Tsuji, Fuyuki Yoshikane
Terminology, 6, 2, 151, 173, 2000, [Peer-reviewed]
English, Scientific journal, This article provides basic background information on the articles included in this special issue on Japanese term extraction, by (i) clarifying the basic background of research into automatic term recognition, (ii) explaining briefl y the ‘contest’-style workshop we organised in 1999, and (iii) briefl y summarising the ATR methodologies proposed in the articles, and positioning their ideas, philosophies and methodologies within ATR from a unifi ed perspective. Through this information, we intend to consolidate the contributions of the NTCIR TMREC workshop, and, hopefully, clarify a basic framework for discussion which different researchers can use to constructively communicate with each other about automatic term extraction and beyond. © 2000 John Benjamins Publishing Co. - New development of NACSIS-IR (New wave of Information Retrieval)
KIMURA Masaru, YOSHIOKA Masaharu, KANDO Noriko, KAGEURA Kyo, OYAMA Keizo
The Journal of Information Science and Technology Association, 50, 1, 30, 36, Information Science and Technology Association, Japan, 2000
Japanese, This paper describes the transformation of NACSIS information retrieval service (NACSIS-IR). NACSIS has provided database search service called NACSIS-IR since 1987. NACSIS-IR plans to start new information retrieval service based on fulltext search engine in January 2000. The authors introduce structure of databases, search functions, WWW-based user interfaces, and other features of new NACSIS-IR. - The Development of Knowledge Integrated Engineering Framework Based on Ontology ("Ontology : Foundations and Applications")
SEKIYA Takayuki, YOSHIOKA Masaharu, TOMIYAMA Tetsuo, Takayuki Sekiya, Masaharu Yoshioka, Tetsuo Tomiyama
Journal of Japanese Society for Artificial Intelligence, 14, 6, 1051, 1060, The Japanese Society for Artificial Intelligence, 01 Nov. 1999
Japanese, During design, a designer uses various computational tools, such as a geometric modeling system, analysis tools, and databases. To support these design processes, a system that can integrate such computational tools is required. For this purpose, we have been developing the Knowledge Intensive Engineering Framework (KIEF) system. KIEF enables the designer to build, evaluate, and modify design models on the system by providing him/her with a wide variety of engineering knowledge. A common ontology plays an important role in integrating tools and representing engineering knowledge. This paper describes the structure of the ontology in KIEF, and how it is used in a modeling processes on KIEF. - Knowledge acpuisition by documenting design for the knowledge intensive engineering framework
NOMAGUCHI Y.
Proceedings of the 1999 Lancaster Internationla Workshop on Engineering Design, CACD'99, 135, 163, May 1999, [Peer-reviewed]
English, International conference proceedings - 設計文書からの知識の構造化の手法の提案
野間口大, 吉岡真治, 冨山哲男
日本機械学会第9回設計工学・システム部門講演会論文集, Apr. 1999
Japanese, Research society - A proposal of design methodology for upgradable products
Yasushi Umeda, Yoshiki Shimomura, Masaharu Yoshioka, Tetsuo Tomiyama
Proceedings of the ASME Design Engineering Technical Conference, 4, 407, 417, American Society of Mechanical Engineers (ASME), 1999
English, International conference proceedings, One of main causes of the environmental issues is the current mass production paradigm. Developing upgradable products is one of the most hopeful approaches for solving these issues, because they will have longer lives, encourage reuse of artifacts, and therefore, create new business chances at later stage.s of product life cycles. However, design for upgradability has been hardly discussed. Thi.s paper proposes a design methodology for upgradable products. First, we clarify the definition of design for upgradability by comparing it with other design strategies. Second, by analyzing data of existing serie.s design, we propose several design pattem.s and two basic guiding principles
namely, making function independent and making function insensitive. Based on these analyses, a design methodology for upgradability is proposed and a prototype of upgrade design support system i.s illustrated. Finally, feasibility of the methodology i.s discussed by applying result of an experimental design to the prototype system. - Toward a reasoning framework of design as synthesis
Masaharu Yoshioka, Tetsuo Tomiyama
Proceedings of the ASME Design Engineering Technical Conference, 3, 1, 11, American Society of Mechanical Engineers (ASME), 1999
English, International conference proceedings, Most of the previous research efforts for design process modeling had such assumptions as "design as problem solving,""design as decision making,"and "design by analysis,"and did not explicitly address "design as synthesis."These views lack notion and understanding about synthesis. Compared with analysis, synthesis is less understood and clarified. This paper discusses our fundamental view on synthesis and approach toward a reasoning framework of design as synthesis. To do so, we observe the designer's activity and formalize knowledge operations in design processes. From the observation, we propose a hypothetical reasoning framework of design based on multiple model-based reasoning. We discuss the implementation strategy for the framework. - Overview of TMREC Tasks.
Kyo Kageura, Masaharu Yoshioka, Koichi Takeuchi, Teruo Koyama, Keita Tsuji, Fuyuki Yoshikane, Maho Okada
Proceedings of the First NTCIR Workshop on Research in Japanese Text Retrieval and Term Recognition, NTCIR-1, Tokyo, Japan, August 30 - September 1, 1999, National Center for Science Information Systems (NACSIS), 1999, [Peer-reviewed] - Evaluation of the Term Recognition Task.
Kyo Kageura, Masaharu Yoshioka, Keita Tsuji, Fuyuki Yoshikane, Koichi Takeuchi, Teruo Koyama
Proceedings of the First NTCIR Workshop on Research in Japanese Text Retrieval and Term Recognition, NTCIR-1, Tokyo, Japan, August 30 - September 1, 1999, 417, 434, National Center for Science Information Systems (NACSIS), 1999, [Peer-reviewed] - Evaluation of the Role Analysis Task.
Teruo Koyama, Masaharu Yoshioka, Koichi Takeuchi, Kyo Kageura
Proceedings of the First NTCIR Workshop on Research in Japanese Text Retrieval and Term Recognition, NTCIR-1, Tokyo, Japan, August 30 - September 1, 1999, National Center for Science Information Systems (NACSIS), 1999, [Peer-reviewed] - Evaluation of the Keyword Extraction Task.
Koichi Takeuchi, Masaharu Yoshioka, Teruo Koyama, Kyo Kageura
Proceedings of the First NTCIR Workshop on Research in Japanese Text Retrieval and Term Recognition, NTCIR-1, Tokyo, Japan, August 30 - September 1, 1999, National Center for Science Information Systems (NACSIS), 1999, [Peer-reviewed] - 設計知識に注目した設計の推論モデルの提案(第2報) --設計行為に関する知識の提案--
武田英明, 吉岡真治, 冨山哲男
日本機械学会 第9回 設計工学・システム部門講演会, 82, 85, Jan. 1999
Japanese - Phrase processing methods for Japanese text retrieval
Kando, N, Kageura, K, Yoshioka, M, Oyama, K
Proceedings of ACM-SIGIR Workshop on Information Retrieval: Theory into Practice. August 28, 1998, Melbourne, Australia, 13, 19, 28 Aug. 1998, [Peer-reviewed]
English, International conference proceedings - A Study on an Integrated Modeling Environment for Design: A Proposal of the Pluggable Metamodel Mechanism
YOSHIOKA Masaharu, TOMIYAMA Tetsuo, Masaharu Yoshioka, Tetsuo Tomiyama
Journal of Japanese Society for Artificial Intelligence, 13, 2, 312, 319, The Japanese Society for Artificial Intelligence, 01 Mar. 1998
Japanese, In this study, we propose a framework of an integrated modeling environment for design. Since a mechanical engineering design process requires various kinds of design object models, such as a geometric model, a kinematic model, and a finite element model, this environment should maintain consistency among these models. Furthermore, commercial design tools are practically used in design processes, and these must be integrated as well. For dealing with multiple design object models, we have formalized a concept called a metamodel. A metamodel is a model which represents relationships among concepts used in various design object models. This paper proposes a framework of the pluggable metamodel mechanism that allows to plug in existing design tools, to support a modeling on it, and to maintain consistency among them. We describe a prototype system and illustrate an example of ball screw design. - Phrase processing methods for Japanese text retrieval
Noriko Kando, Kyo Kageura, Masaharu Yoshioka, Keizo Oyama
SIGIR Forum (ACM Special Interest Group on Information Retrieval), 32, 2, 23, 28, Association for Computing Machinery (ACM), 1998, [Peer-reviewed]
English, Scientific journal, This paper examines the effectiveness of different phrase identification and weighting methods for Japanese text retrieval in an operational information retrieval (IR) system, called NACSIS-IR. Based on our previous experiments, we used character-based indexing with positional information and word-or phrase-based query processing, which allowed us to implement sophisticated linguistic analysis on large-scale databases while maintaining adequate efficiency. The results of retrieval experiments on a large-scale Japanese test collection showed that the combination of enhanced phrase identification using patterns defined over part-of-speech tags and our algorithms Phrase2 and Phrase5 made a significant positive contribution to retrieval effectiveness. The paper also discusses indexing and phrase processing of Japanese or East Asian languages. - Representation of design object based on functional evolution process model
Yoshiki Shimomura, Masaharu Yoshioka, Hideaki Takeda, Yasushi Umeda, Tetsuo Tomiyama
Journal of Mechanical Design, 120, 2, 221, 229, Jan. 1998, [Peer-reviewed]
English, Scientific journal - NTCIR : NACSIS Test Collection Project
Noriko Kando, Teruo Koyama, Keizo Oyama, Kyo Kageura, Masaharu Yoshioka
Proceedings of the 20th Annual Colloquium of BCS-IRSG, Jan. 1997, [Peer-reviewed]
English, International conference proceedings - The modeling of synthesis -from the viewpoint of design knowledge
T. Tomiyama, T. Murakami, T. Washio, A. Kubota, H. Takeda, T. Kiriyama, Y. Umeda, M. Yoshioka
WDK 25, Proceedings of the 11th International Conference on Engineering Design in Tampere 199, Tampere, Finland, 1997., 3, 97, 100, Jan. 1997, [Peer-reviewed]
English, International conference proceedings - 複数の設計対象モデルを扱う設計過程モデル
吉岡真治, 中村雅彦, 冨山哲男, 吉川弘之
日本機械学會論文集. C編, 61, 581, 330, 337, 社団法人日本機械学会, Dec. 1995, [Peer-reviewed]
Scientific journal, An intelligent CAD system aims at serving a designer as an intelligent, integrated design environment. For building such a system, development of a large scale knowledge base is necessary. To do so, design knowledge must be systematically formalized, made computable, and organized. Design knowledge has two categories ; i.e., design object knowledge and design process knowledge. In our previous reports, we have formalized a computable design process model to represent design process knowledge and the concept of a metamodel dealing with multiple design object models. This paper proposes a fra... - Representation of design object based on the functional evolution process model
Yoshiki Shimomura, Hideaki Takeda, Masaharu Yoshioka, Yasushi Umeda, Tetsuo Tomiyama
American Society of Mechanical Engineers, Design Engineering Division (Publication) DE, 83, 2 Pt 1, 351, 360, 1995, [Peer-reviewed]
English, International conference proceedings, An extension of the Function-Behavior-State (FBS) diagram is proposed for representing a function in the computer aided conceptual design process. The new FBS method uses three additional concepts, namely: function body, function modifier and object entity, to represent the designers' intention more precisely than the original FBS diagram. The performance of the proposed method is complemented by the Functional Evolution Process (FEP) model which represents the different design processes. - Design process model with multiple design object models
Masaharu Yoshioka, Masahiko Nakamura, Tetsuo Tomiyama, Hiroyuki Yoshikawa
American Society of Mechanical Engineers, Design Engineering Division (Publication) DE, 53, 7, 14, 1993, [Peer-reviewed]
International conference proceedings, An intelligent CAD system is not merely a set of intelligent design tools, but rather it must be an intelligent, integrated design environment. This requests it must be equipped with a large scale knowledge base in which design knowledge is intensively and systematically stored. To do so, design knowledge must be systematically formalized, made computable, and organized. The present paper investigates fundamental issues of systematization of design knowledge. Design knowledge has two categories; i.e., design object knowledge and design process knowledge. In our previous reports, we have formalized a computable design process model to represent design process knowledge and the concept of metamodel dealing with multiple design object models. This paper proposes a framework to achieve harmonized combination of these two. In this framework, the computable design process model navigates the metamodel mechanism to generate and modify models of the design object. We describe a prototype system and illustrate an example of ship design. - 多重世界を用いた設計過程の論理による形式化
武田英明, 吉岡真治, 冨山哲男, 吉川弘之
1991年度人工知能学会(第5回)論文集, 学習院大学、東京, 1991. 1991年6月28日, 28 Jun. 1991
Japanese - 多重世界を用いた設計過程のシミュレーション
吉岡真治, 河合浩之, 武田英明, 冨山哲男, 吉川弘之
1991年度精密工学会春季大会講演論文集, 早稲田大学、東京, 1991. 1991年3月26日発表, 26 Mar. 1991
Japanese
Other Activities and Achievements
- Preface of Special Issue on 10th Competition on Legal Information of Extraction and Entailment (COLIEE 2023).
Randy Goebel, Yoshinobu Kano, Mi-Young Kim, Juliano Rabelo 0001, Ken Satoh, Masaharu Yoshioka, Rev. Socionetwork Strateg., 18, 1, 25, 26, Apr. 2024 - 特定の専門分野を対象とした意味役割付きデータ作成手法~有機合成手順の抽出を例として~
町光二郎, 秋山世治, 長田裕也, 吉岡真治, 吉岡真治, 吉岡真治, 言語処理学会年次大会発表論文集(Web), 30th, 2024 - Proposal for an elementary reaction knowledge base for interpreting computational results of chemical reaction processes
小嶋陽平, 小野ゆり子, NATH Pinku, 吉岡真治, 吉岡真治, 吉岡真治, 人工知能学会全国大会論文集(Web), 38th, 2024 - Pseudo Training Data Generation for Automatic Aggregation of Open-Ended Questionnaire Responses by Large Language Models
長谷川遼, 銭本友樹, 宇津呂武仁, 西崎博光, 吉岡真治, 神門典子, 人工知能学会全国大会論文集(Web), 38th, 2024 - Proposal of qualitative reasoning system that has more compatibilitywith large language model based on physical laws and principles.
YOSHIOKA Masaharu, SUZUKI Shinpei, Proceedings of the Annual Conference of JSAI, JSAI2024, 3Xin228, 3Xin228, 2024
Recent development of Large Language Model (LLM) allows the user to obtain the explanation of physical phenomenon as text. However, the output of LLM may contain inappropriate explanation, and it is not easy to justify the appropriateness of the generated text. To solve this problem, we have proposed a framework to construct a model for qualitative reasoning system based on the text analysis results of LLM (e.g., a list of objects and physical phenomena). In this paper, we propose a qualitative reasoning system that has more compatibility to interpret the output of LLM and demonstrate the system with simple physical behavior simulation., The Japanese Society for Artificial Intelligence, Japanese - Model construction method for qualitative reasoning based on the recognition results of physical laws and principles by large language models
SUZUKI Shinpei, YOSHIOKA Masaharu, Proceedings of the Annual Conference of JSAI, JSAI2024, 3Xin227, 3Xin227, 2024
The Large Language Model (LLM) has recently been developed and has demonstrated impressive performance in solving a variety of tasks. However, LLMs still face challenges when solving tasks that require the combination of multiple rules simultaneously. For instance, behavior analysis based on physical laws and principles is one such task. This paper proposes using qualitative reasoning as a framework to combine multiple physical laws and principles for behavior analysis. Furthermore, we propose a method for constructing a model for qualitative reasoning based on the output of LLM. The analysis of a basic physics problem illustrates this method. This method takes advantage of the strengths and compensates for the shortcomings of both LLM and conventional qualitative reasoning model., The Japanese Society for Artificial Intelligence, Japanese - Classification of articles using Wikipedia categories and definition sentences
SUZUKI Nozomi, YOSHIOKA Masaharu, Proceedings of the Annual Conference of JSAI, JSAI2024, 4Xin284, 4Xin284, 2024
There have been several attempts to extract knowledge from Wikipedia. One of the most important information to be extracted is the classification of articles. The Wikipedia category, which classifies the group of articles for ease of navigation, seems to be a good resource for this task. However, since Wikipedia categories are also used for different purposes and many categories are added to an article, it is necessary to select the representative Wikipedia category for better classification. In this paper, we propose to use the definition sentence, which is the first sentence of the article, to select the representative category among them. In this method, we extract the definition word, which is for representing the class of the article, from the definition sentence. We also propose a method to classify the article using this information and categories, and evaluate the method using the SHINRA dataset., The Japanese Society for Artificial Intelligence, Japanese - The Analysis of Wikipedia Category Ontology: Proposal of new category types and analysis using them
EBISUI Yusuke, YOSHIOKA Masaharu, Proceedings of the Annual Conference of JSAI, JSAI2024, 3Xin229, 3Xin229, 2024
Wikipedia is one of the largest online encyclopedias, and there have been attempts to extract structured knowledge from Wikipedia, such as DBpedia, YAGO, and the Japanese Wikipedia Ontology. We have proposed the Wikipedia Category Ontology, which categorizes the types of Wikipedia categories and the relationships between them. In this framework, there are two main types of categories. One is a set category, which represents concepts such as "country" and "university". The other is a topic category, which represents specific topics such as "Japan" and "Hokkaido University". These classifications are decided based on the meaning of the Wikipedia category labels. However, based on the analysis of the actual role of categories in each article, there are certain categories whose articles are not instances of the category. In this paper, we analyze the Wikipedia Category Ontology and propose a new definition and a new method for category classification based on this analysis., The Japanese Society for Artificial Intelligence, Japanese - Report on the 9th ACM SIGIR / the 13th International Conference on the Theory of Information Retrieval (ICTIR 2023).
Masaharu Yoshioka, Mohammad Aliannejadi, Julia Kiseleva, SIGIR Forum, 57, 2, 10, 5, Dec. 2023, [Invited] - A Music Recommendation System that uses Formal Concept Analysis to Reconstruct User Interests from Different Perspectives
椿康平, 吉岡真治, 大久保好章, 人工知能学会人工知能基本問題研究会資料, 124, 07, 12, 06 Mar. 2023
一般社団法人 人工知能学会, Japanese - 学生フォーラム(第115回)吉岡真治先生インタビュー「多分野での経験が後につながる」
吉岡 真治, 人工知能 : 人工知能学会誌 : journal of the Japanese Society for Artificial Intelligence, 38, 1, 81, 84, Jan. 2023
東京 : 人工知能学会 ; 2014-, Japanese - Analysis system for research paper database using chemical reaction Information extraction system
YOSHIOKA Masaharu, MACHI Kojiro, NAGATA Yuuya, Proceedings of the Annual Conference of JSAI, JSAI2023, 4Xin145, 4Xin145, 2023
We have been working on the project to develop a new research paper database that collects papers from the particular domain and extracts important terms to analyze the cooccurrence among them. This system has been applied to the research domain of nanocrystal device development for evaluation. In this paper, to discuss the generality of the proposed system, we have applied the system to the domain of organic chemical reaction development, which is similar to the previous domain. In this domain, there are open databases for representing the chemical substance and information extraction tool. These tools are good for making the initial database, but we confirm it is also necessary to consider the variation of representing the same chemical substance in different string., The Japanese Society for Artificial Intelligence, Japanese - Preface of Special Issue on 8th Competition on Legal Information of Extraction and Entailment (COLIEE 2021).
Randy Goebel, Yoshinobu Kano, Mi-Young Kim, Juliano Rabelo 0001, Ken Satoh, Masaharu Yoshioka, The Review of Socionetwork Strategies, 16, 1, 109, 110, 2022 - Extended Overview of ChEMU 2022 Evaluation Campaign: Information Extraction in Chemical Patents.
Yuan Li, Biaoyan Fang, Jiayuan He 0002, Hiyori Yoshikawa, Saber A. Akhondi, Christian Druckenbrodt, Camilo Thorne, Zubair Afzal, Zenan Zhai, Kojiro Machi, Masaharu Yoshioka, Youngrok Jang, Hosung Song, Junho Lee, Gyeonghun Kim, Yireun Kim, Stanley Jungkyu Choi, Honglak Lee, Kyunghoon Bae, Darshini Mahendran, Christina Tang, Bridget T. McInnes, Timothy Baldwin, Karin Verspoor, CLEF (Working Notes), 758, 781, 2022
CEUR-WS.org - Wikipedia Category Ontology - Utilization of Wikipedia category from the view point of knowledge engineers -
YOSHIOKA Masaharu, JSAI Technical Report, Type 2 SIG, 2021, 55, 5, 5, 2021Wikipedia is one of the largest knowledge resources available via the Internet. Several methods have been proposed for extracting knowledge/information from Wikipedia, but methods based on extracting knowledge from the Wikipedia category structure can not utilize whole part of its structure because of the complexity of the relationships between the various categories. In this paper, we briefly review our previous researches on analyzing Wikipedia category and introduce a framework called "Wikipedia Category Ontology" (WCO) that aims to act as a basis for interpreting the Wikipedia category structure. It is based on a classification of category types and relationship types and available online in the form of Linked Open Data at http://wcontology.org/. We also demonstrate the system by using Wikipedia category analysis scenarios.
, The Japanese Society for Artificial Intelligence, Japanese - NTCIR15QA Lab-PoliInfo-2の報告およびデータセット公開
木村泰知, 渋木英潔, 高丸圭一, 内田ゆず, 乙武北斗, 石下円香, 三田村照子, 吉岡真治, 秋葉友良, 小川泰弘, 佐々木稔, 横手健一, 神門典子, 森辰則, 荒木健治, 関根聡, 言語処理学会年次大会発表論文集(Web), 27th, 2021 - 日本語Wikipediaカテゴリを用いた記事のクラス分類手法の提案
小板橋佳晃, 吉岡真治, 吉岡真治, 言語処理学会年次大会発表論文集(Web), 27th, 2021 - A Proposal of a Method for Enumerating and Indexing Set Partitions Using ZDDs and Its Experimental Evaluation
高橋翔哉, 吉岡真治, 吉岡真治, 人工知能学会人工知能基本問題研究会資料, 112, 42, 47, 08 Mar. 2020
人工知能学会, Japanese - Introduction of Pruning Using Graph Neural Networks to Algorithms for Combinatorial Optimization Problems on Graphs
中野裕太, 吉岡真治, 吉岡真治, 吉岡真治, 吉岡真治, 人工知能学会人工知能基本問題研究会資料, 111, 23, 28, 29 Jan. 2020
人工知能学会, Japanese - NTCIR-15QA Lab-PoliInfo2のタスク設計
木村泰知, 渋木英潔, 高丸圭一, 秋葉友良, 石下円香, 内田ゆず, 小川泰弘, 乙武北斗, 佐々木稔, 三田村照子, 横手健一, 吉岡真治, 神門典子, 神門典子, 言語処理学会年次大会発表論文集(Web), 26th, 2020 - 森羅と日本版DBpediaにおける属性の取り扱いに関する比較分析
小板橋佳晃, 吉岡真治, 吉岡真治, 言語処理学会年次大会発表論文集(Web), 26th, 2020 - ユーザの興味を反映したニュースサイトの多観点スタンス分析
立浪紀彦, 吉岡真治, 神門典子, ALLAN James, 伊藤正彦, 言語処理学会年次大会発表論文集(Web), 26th, 2020 - 無機化合物を対象とした論文に対する化学物質名抽出システムの性能分析
町光二郎, 吉岡真治, 吉岡真治, 吉岡真治, 吉岡真治, 言語処理学会年次大会発表論文集(Web), 26th, 2020 - Experimental Analysis of Chemical Named Entity Recognition System Using Chemical Database for Training Data
町光二郎, 吉岡真治, 情報処理学会研究報告(Web), 2020, IFAT-139, 2020 - Classification of Wikipedia pages using Wikipedia category ontology
NAKAGAWA Takanori, KOITABASHI Yoshiaki, YOSHIOKA Masaharu, Proceedings of the Annual Conference of JSAI, 2020, 0, 4H3GS405, 4H3GS405, 2020Since a category of Wikipedia contains information useful for classifying a Wikipedia pages. A classification method using this category information has been proposed.However, since various types of information are mixed in the Wikipedia category, there is a problem that a category that is not appropriate to be assigned at the same time is added.In this research, we classify more appropriate classes for pages with multiple categories using Wikipedia category ontology, which defines the types of Wikipedia categories and their parent-child relationships. By using this method, we rank the adequacy of the page class belonging to a plurality of categories. To confirm effectiveness of this method, we compar with the classification of SHINRA : Wikipedia Structuring Project.
, The Japanese Society for Artificial Intelligence, Japanese - Analysis of the Statements Expressing the Pros and Cons of a Bill in Tokyo Metropolitan Assembly:Toward NTCIR-15 QA Lab-PoliInfo-2 Stance Classification Task
TAKAMARU Keiichi, KIMURA Yasutomo, UCHIDA Yuzu, SASAKI Minoru, YOSHIOKA Masaharu, AKIBA Tomoyoshi, SHIBUKI Hideyuki, Proceedings of the Annual Conference of JSAI, 2020, 0, 4Q3GS901, 4Q3GS901, 2020In the NTCIR-15 QA Lab-PoliInfo-2, we are planning a Stance Classification Task that estimates the pros and cons of each bill of the group in the Tokyo Metropolitan Assembly. This paper describes the outline of the task and the construction of the data set for the task. Then, this paper analyzes the existence of remarks and their expression forms for expressing approval or disapproval of bills included in the minutes of the Tokyo Metropolitan Assembly in 2018, which is part of the data used in the task.
, The Japanese Society for Artificial Intelligence, Japanese - Feasibility study for technical documents retrieval system in construction industry
FURUKAWA Kei, YOSHIOKA Masaharu, Proceedings of the Annual Conference of JSAI, 2020, 0, 3Rin470, 3Rin470, 2020Lots of technical documents have been created and accumulated for the purpose of standardizing and inheriting technologies in many companies across industries. Those documents have certain formats and can be searched by documents retrieval systems. However, since many of those documents are written in specific fields, it's difficult for younger engineers to find the documents related to their tasks if they don't know appropriate key words whereas it's easy for experienced engineers who know the existence of such documents. In this study, we conduct feasibility studies to reveal the above problems with documents in a construction company as an example and introduce the basic idea of the new technical documents retrieval system utilizing construction steps and conditions.
, The Japanese Society for Artificial Intelligence, Japanese - Market trend prediction using multiple technical indicators
KATAYOSE Ryosuke, MASAHARU Yoshioka, Proceedings of the Annual Conference of JSAI, 2020, 0, 3Rin412, 3Rin412, 2020Stock price forecasting is performed in various forms, such as technical analysis and fundamental analysis using time series data and financial indicators, and text analysis using news information. We propose a method for predicting stock prices by machine learning using multiple technical indicators, taking into account the increasing number of algorithmic trades using technical analysis. At this time, stocks with good and bad fits by technical analysis are classified and evaluated. We consider how to combine multiple technical indicators.
, The Japanese Society for Artificial Intelligence, Japanese - 専門用語の知識保全エコシステムを有するインハウス論文・図表データベースの構築
吉岡真治, 尹磊, 原真二郎, 鈴木晃, 高山英紀, 石井真史, 言語処理学会年次大会発表論文集(Web), 25th, 2019 - 法律文書処理の国際コンテスト型ワークショップCOLIEE2018の開催
狩野芳伸, 吉岡真治, KIM Mi-Young, LU Yao, RABELO Juliano, 清田直希, GOEBEL Randy, 佐藤健, 言語処理学会年次大会発表論文集(Web), 25th, 2019 - Wikipediaカテゴリの構成要素に注目したカテゴリ階層の分析
中川嵩教, 小板橋佳晃, 吉岡真治, 吉岡真治, 言語処理学会年次大会発表論文集(Web), 25th, 2019 - 更新可能な用語抽出機能を持つ小規模研究グループ向け論文・図表データベースの構築
吉岡真治, 吉岡真治, 大久保好章, 尹磊, 原真二郎, 鈴木晃, 高山英紀, 石井真史, 応用物理学会秋季学術講演会講演予稿集(CD-ROM), 80th, 2019 - 知識工学者のためのWikipediaカテゴリの分析とその活用
吉岡真治, 吉岡真治, 人工知能学会人工知能基本問題研究会資料, 110th, 2019 - Reorganizing Hierarchical Category Structure of Wikipedia Based on the Parent-Child Relationship Classification
NAKAGAWA Takanori, KOITABASHI Yoshiaki, YOSHIOKA Masaharu, Proceedings of the Annual Conference of JSAI, 2019, 0, 1K2J405, 1K2J405, 2019
<p>Wikipediaカテゴリを分析する研究から、カテゴリには概念分類を表すものだけでなく、人名や組織名といったインスタンスを表すものやその組み合わせが存在しているために、知識工学的観点から利用するためには、そのカテゴリ間の関係について検討することが必要であることが提案された。また、この考え方に基づき、カテゴリの種類、カテゴリの親子関係の種類を整理したデータの構築を行った。本稿では、この構築したデータに基づき、特定の親子関係のみを利用して(あるいは、排除して)できあがるカテゴリ階層構造の性質を議論することで、これまでに提案してきたWikipediaカテゴリに関する分類の枠組の有用性について議論する。</p>, The Japanese Society for Artificial Intelligence, Japanese - 大規模ニュース記事データベースを用いた複数観点からの記事の対比と報道スタンスの分析
立浪紀彦, 吉岡真治, 情報処理学会研究報告(Web), 2018, ICS-192, Vol.2018‐ICS‐192,No.6,1‐4 (WEB ONLY), 29 Jun. 2018
Japanese - 知識工学者のためのWikipediaカテゴリ構造の再整理の構想
吉岡真治, 中川嵩教, 言語処理学会年次大会発表論文集(Web), 24th, ROMBUNNO.P12‐1 (WEB ONLY), 2018
Japanese - Reorganizing Hierarchical Category Structure of Japanese Wikipedia for Knowledge Engineers
NAKAGAWA Takanori, YOSHIOKA Masaharu, Proceedings of the Annual Conference of JSAI, 2018, 0, 2F402, 2F402, 2018Since Wikipedia is an on-line encyclopedia that covers varieties topics, there are several attempts to extracting knowledge from its contents. Wikipedia category is one type of information that provides topical index for the articles and organized as hierarchical manner. However, since hierarchical structure of Wikipedia category have different characteristics compared to the one for knowledge representation, it may not fully and correctly utilized at this moment. In this paper, we propose a method to extract hierarchical structure that have similar characteristics with knowledge representation.
, The Japanese Society for Artificial Intelligence, Japanese - 多観点分析に基づくトピック推定と特許マップへの応用
小野寺大輝, 吉岡真治, 言語処理学会年次大会発表論文集(Web), 23rd, ROMBUNNO.C3‐4 (WEB ONLY), 2017
Japanese - 人工知能の製造物責任とリスクに関する試論
吉岡 真治, 人工知能学会全国大会論文集, 2017, 0, 1E1OS24a3, 1E1OS24a3, 2017本試論では、自動運転の車のようなオープンな環境で利用される人工知能を搭載された人工物の製造物責任について検討する。オープンな環境における責任を議論するためには、様々な事態と付随するリスクを認識し、製造者と利用者が適切に分担する枠組が必要になると考えている。本論では、トロッコ問題を例にとり、製造者と利用者がリスクを適切に認識するための枠組とその実現のための人工知能の果たすべき役割について考察する。
, 一般社団法人 人工知能学会, Japanese - NaDev: An Annotated Corpus to Support Information Extraction from Research Papers on Nanocrystal Devices
Thaer M. Dieb, Masaharu Yoshioka, Shinjiro Hara, 情報処理学会論文誌, 57, 4, 15 Apr. 2016
The process of nanocrystal device development is not well systematized. To support this process, analysis of the information produced by developmental experiments is required. In this study, we constructed an annotated corpus to support the extraction of experimental information from relevant publications. We designed the corpus-construction guidelines by cooperating with a domain expert. We evaluated these guidelines through corpus-construction experiments with graduate students from this domain, and then evaluated the corpus with the domain expert. In the corpus construction experiments, we achieved a sufficient level of Inter-Annotator Agreement by using a loose agreement measure that ignored the term-boundary mismatch problem, and made an agreement corpus that excluded annotations based on misunderstanding the guidelines. The domain expert evaluated this agreement corpus and modified the guidelines based on real examples. Using these guidelines, we finalized the corpus called "NaDev" (Nanocrystal Device development corpus). The NaDev corpus and its construction guidelines will be released via our website, http://nanoinfo.ist.hokudai.ac.jp/. The NaDev corpus aims to support automatic information extraction from publications relevant to nanocrystal device development. This information can be used to solve problems in the nanotechnology domain using the massive availability of fresh information. To the best of our knowledge, this is the first corpus constructed for the development of nanocrystal devices.\n------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.24(2016) No.3 (online)------------------------------The process of nanocrystal device development is not well systematized. To support this process, analysis of the information produced by developmental experiments is required. In this study, we constructed an annotated corpus to support the extraction of experimental information from relevant publications. We designed the corpus-construction guidelines by cooperating with a domain expert. We evaluated these guidelines through corpus-construction experiments with graduate students from this domain, and then evaluated the corpus with the domain expert. In the corpus construction experiments, we achieved a sufficient level of Inter-Annotator Agreement by using a loose agreement measure that ignored the term-boundary mismatch problem, and made an agreement corpus that excluded annotations based on misunderstanding the guidelines. The domain expert evaluated this agreement corpus and modified the guidelines based on real examples. Using these guidelines, we finalized the corpus called "NaDev" (Nanocrystal Device development corpus). The NaDev corpus and its construction guidelines will be released via our website, http://nanoinfo.ist.hokudai.ac.jp/. The NaDev corpus aims to support automatic information extraction from publications relevant to nanocrystal device development. This information can be used to solve problems in the nanotechnology domain using the massive availability of fresh information. To the best of our knowledge, this is the first corpus constructed for the development of nanocrystal devices.\n------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.24(2016) No.3 (online)------------------------------, English - 対象‐観点を考慮したfacet‐biasedトピックモデルと特許マップへの応用
小野寺大輝, 吉岡真治, 言語処理学会年次大会発表論文集(Web), 22nd, P13‐5 (WEB ONLY), 2016
Japanese - facet‐biasedトピックモデルと距離尺度学習を用いたニュース記事の分類
小野寺大輝, HUANG Le, 吉岡真治, 人工知能学会全国大会論文集(CD-ROM), 2016, 0, 4B45, 4B45, 2016
一般社団法人 人工知能学会, Japanese - ナノ知識探索プロジェクト:実験記録からの知識発見(第4報)―キャプションからのメタデータの自動抽出によるグラフイメージ検索システム―
ZHU Tao, DIEB Thaer M, 吉岡真治, 原真二郎, 人工知能学会全国大会論文集(CD-ROM), 30th, ROMBUNNO.1J2‐4, 2016
Japanese - ナノ知識探索プロジェクト:実験記録からの知識発見(第4報)
朱 濤, Dieb Thaer Mostafa, 吉岡 真治, 原 真二郎, 人工知能学会全国大会論文集, 2016, 0, 1J24, 1J24, 2016
一般社団法人 人工知能学会, Japanese - DBPediaの情報に基づくWikipediaのカテゴリ情報の一貫性の分析
吉岡 真治, Loban Rhett, 人工知能学会全国大会論文集, 2015, 0, 1G42, 1G42, 2015Web上の百科事典であるWikipediaは、様々な事象に関する知識源であるが、その情報の品質については、編集者に依存する。本研究では、DBPediaにより抽出された構造化情報を用いて、Wikipediaのカテゴリ構造の一貫性を分析する方法を提案するとともに、その分析事例について紹介する。
, 一般社団法人 人工知能学会, Japanese - Multifacet comparative analysis of newspaper articles from different conutries - Analysis based on Global Database of Events, Language and Tone (GDELT) - (Awarded JSAI Incentive Award)
Masaharu Yoshioka, Noriko Kando, The Proceedings of the 8th Symposium on JSAI Interactive Visualization and Mining, SIG-AM-08-0, 24, 31, Nov. 2014
Japanese - 特長表現に注目した対象-観点型特許マップの自動生成
岸桂太, 吉岡真治, 研究報告情報基礎とアクセス技術(IFAT), 2014, 9, 1, 5, 22 Mar. 2014
多くの特許では、既存の製品の機能向上や好ましくない点の抑制といった効果を目指している。これらの効果に関する記述は、定型的に表現されることが多く、これを特長表現と呼ぶ。本研究では、特長表現から対象と観点を抽出し、それらを用いた特許マップの生成手法を提案する。, Japanese - Wikipediaのカテゴリー階層関係の分類を用いた日本語Wikipediaオントロジーの分析
吉岡 真治, 人工知能学会全国大会論文集, 2014, 0, 2J34, 2J34, 2014Wikipediaのカテゴリーは、閲覧者にとって有用な分類を提供するものであり、そのカテゴリーは束状の階層関係として表現される。 この関係には、概念間の階層関係を含む様々な関係が存在する。 本発表では、このカテゴリー階層関係の特徴的なパターンを分析すると共に、その分析結果に基づいた日本語Wikipediaオントロジーに関する考察を行う。
, 一般社団法人 人工知能学会, Japanese - 特長表現に注目した特許マップの自動生成 (データ工学)
岸 桂太, 吉岡 真治, 電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 113, 150, 55, 58, 22 Jul. 2013
多くの特許では、既存の製品の機能向上や好ましくない点の抑制といった効果を目指している。これらの効果に関する記述は、定型的に表現されることが多く、これを特長表現と呼ぶ。本研究では、特長表現を網羅的に集めるための手法を提案するとともに、効果とその対象に注目した特許マップの生成手法を提案する。, 一般社団法人電子情報通信学会, Japanese - ナノ知識探索プロジェクト:実験記録からの知識発見 (第3報) – Nanodevice Research Papers Clustering Based on Automatic Paper Annotation –
Thaer Moustafa Dieb, 吉岡 真治, 原 真二郎, 2013年度(第27回)人工知能学会全国大会 (JSAI 2013), 富山, 2013年6月4日-7日, 発表論文集, 1C3-4 (4pp.), 1C3-4-1, 1C3-4-4, Jun. 2013
English - Diversity of Topics in Collecting Web Pages using a Search Engine AP
Daichi KOIKE, KensakuMAKITA, TakehitoUTSURO, Yasuhide KAWADA, Masaharu YOSHIOKA, Noriko KANDO, Proceedings of the 5th Forum for Data Engineering and Information Management, Mar. 2013
Japanese - ニュース・ツイッター間の対応を考慮したバースト・トピックの同定
高橋 佑介, 胡 碩, 宇津呂 武仁, 吉岡 真治, 神門 典子, 第5回データ工学と情報マネジメントに関するフォーラム---DEIM フォーラム--- 論文集, Mar. 2013
Japanese - 日中時系列ニュースにおけるバースト・トピックの推定と二言語間対応付け
HU Shuo, 高橋佑介, ZHENG LiYi, 宇津呂武仁, 吉岡真治, 神門典子, 言語処理学会年次大会発表論文集(Web), 19th, WEB ONLY C3-1, 2013
Japanese - 時系列中国語ニュース・ブログにおけるトピックモデルの推定と比較対照分析
ZHENG Liyi, HU Shuo, 小池大地, 宇津呂武仁, 吉岡真治, 神門典子, 言語処理学会年次大会発表論文集(Web), 19th, WEB ONLY C3-3, 2013
Japanese - 検索エンジンAPIを用いて収集したウェブページにおけるトピック多様性の分析
小池大地, 宇津呂武仁, 河田容英, 吉岡真治, 人工知能学会全国大会論文集(CD-ROM), 2013, 0, 2E14in, 2E14in, 2013本論文では,検索エンジンAPIを用いたウェブページ収集タスクにおいて,ト ピックモデルを用いたトピックの分布の観点から,できる限り多様なウェブペー ジを収集する方式を提案する.提案方式によって,ニュース・ブログにおける トピックとの関連が強いウェブページを収集することにより,より多様性に富 んだトピックの分布が観測できることを示す.
, 一般社団法人 人工知能学会, Japanese - ナノ知識探索プロジェクト:実験記録からの知識発見(第3報)
Thaer Mostafa Dieb, 吉岡 真治, 原 真二郎, 人工知能学会全国大会論文集, 2013, 0, 1C34, 1C34, 2013We have been working on the project "Exploring knowledge for nanodevice development" and proposed a framework to annotate useful information (e.g., source material, evaluation parameter, and so on) for analyzing nanocrystal device development papers. In this paper, we conduct nanocrystal device research papers clustering experiment once on annotated papers using automatic annotation results, and once as non-annotated using bag of word approach, and then we compare the results discussing the usefulness of the automatic annotation.
, 一般社団法人 人工知能学会, Japanese - ナノ知識探索プロジェクト―ナノデバイス開発記録からの知識発見―
吉岡真治, THAER Moustafa Dieb, 原真二郎, Designシンポジウム講演論文集, 2012, 87, 91, 16 Oct. 2012
Japanese - ナノ知識探索プロジェクト:ナノデバイス開発記録からの知識発見
吉岡 真治, Thaer Moustafa Dieb, 原 真二郎, Designシンポジウム2012, 京都, 2012年10月16日-17日, 発表論文集 (5pp.), 1, 5, Oct. 2012
Japanese - Discovery and Maintenance of Links between Wikipedia and GeoNames by Using Wikipedia Category
吉岡 真治, 劉 亦奇, 神門 典子, 情報処理学会論文誌データベース(TOD), 5, 3, 141, 148, 28 Sep. 2012
Recently, due to the higher demand for geographic information system, it is necessary to have a good geographical database for such systems. GeoNames is one of the largest geographical database as Open Data. This database is also used for constructing web ontology by adding links to the Wikipedia page as a part of Linked Open Data. Even though, here are several attempts to find links automatically, the number of links between GeoNames and Wikipedia is not sufficient. In this paper, we propose an automatic link discovery method to use Wikipedia categories to identify the correspondence between Wikipedia page and GeoNames entry. We also propose to use this method for inappropriate link detection. Link data obtained in this paper is already sent to the manager of GeoNames and a part of the result is used for updating the site., 情報処理学会, Japanese - Estimation and Analysis of Topic Models in Time Series Japanese / Chinese News and Blogs
HU Shuo, ZHENG Liyi, TAKAHASHI Yusuke, KOIKE Daichi, MAKITA Kensaku, UTSURO Takehito, YOSHIOKA Masaharu, IEICE technical report. Natural language understanding and models of communication, 112, 196, 25, 30, 23 Aug. 2012
This paper studies issues regarding topic modeling of information flow in multilingual news and blog streams. If someone wants to find differences in the topics of Japanese and Chinese news / blog, it is usually necessary for him/her to carefully watch every article in Japanese and Chinese news streams at every moment. In such a situation, topic models such as LDA (Latent Dirichlet Allocation) and DTM (dynamic topic model) are quite effective in estimating distribution of topics over a document collection such as articles in a news / blog stream. Based on the results of estimating distribution of topics in Japanese / Chinese news streams, this paper proposes how to analyze cross-lingual alignment of topics in time series Japanese / Chinese news streams. This paper also proposes how to analyze differences of concerns and opinions among various Chinese news sites as well as blog hosts., The Institute of Electronics, Information and Communication Engineers, Japanese - Wikipediaを中心としたLinked Open Dataに関する一考察
吉岡真治, 研究報告情報基礎とアクセス技術(IFAT), 2012, 1, 1, 5, 25 Jul. 2012
現在、Wikipedia の情報を中心に様々な Open Data を関連づけて活用する Linked Open Data の研究が行われている。本発表では、Wikipedia のページが表現する情報の粒度が与える影響について考察を行う。, 一般社団法人情報処理学会, Japanese - Summary of NTCIR-9 and A Look Forward
Tetsuya Sakai, Hideo Joho, Noriko Kando, Tsuneaki Kato, Akiko Aizawa, Tomoyoshi Akiba, Isao Goto, Fuminori Kimura, Teruko Mitamura, Hiromitsu Nishizaki, Hideki Shima, Masaharu Yoshioka, Shlomo Geva, Ling-Xiang Tang, Andrew Trotman, Yue Xu, IPSJ SIG Notes, 2012, 5, 1, 15, 19 Mar. 2012
NTCIR is a sesquiannual evaluation workshop series on information access which started in 1999 and is run by National Institute of Informatics (NII). The ninth workshop (NTCIR-9) was concluded in December 2011 and the accepted tasks for the tenth workshop (NTCIR-10) have been announced in January 2012. This paper provides a brief summary of NTCIR-9, and introduces some latest developments from NTCIR-10., 一般社団法人情報処理学会, Japanese - Construction of Tagged Corpus for Nanodevices Development Papers (2ed report): Information Extraction by Cascading Named Entity Recognition Tools
DIEB THAER M, YOSHIOKA MASAHARU, HARA SHINJIRO, 言語処理学会年次大会発表論文集, 18th (CD-ROM), ROMBUNNO.B5-2, 13 Mar. 2012
English - 日中時系列ニュースにおけるトピックの推定と二言語間対応付け
HU Shuo, 高橋佑介, 牧田健作, 横本大輔, 宇津呂武仁, 吉岡真治, 言語処理学会年次大会発表論文集, 18th (CD-ROM), ROMBUNNO.F1-5, 13 Mar. 2012
Japanese - 複数ドメインの意見分析コーパスを用いたアンサンブル学習による意見分析システムの提案
高村慎太郎, 吉岡真治, 関洋平, 言語処理学会年次大会発表論文集, 18th (CD-ROM), ROMBUNNO.P1-11, 13 Mar. 2012
Japanese - 時系列ニュースにおけるトピックのバーストの同定
高橋佑介, 横本大輔, 宇津呂武仁, 吉岡真治, 言語処理学会年次大会発表論文集, 18th (CD-ROM), ROMBUNNO.F1-4, 13 Mar. 2012
Japanese - 震災を題材としたニュース・ブログ間の話題の相関と遷移の分析
小池大地, 横本大輔, 牧田健作, 鈴木浩子, 宇津呂武仁, 河田容英, 吉岡真治, 福原知宏, 言語処理学会年次大会発表論文集, 18th (CD-ROM), ROMBUNNO.A5-1, 13 Mar. 2012
Japanese - Idetification of Burst in the Chronological Topic Model
Yusuke Takahashi, Daisuke Hashimoto, Takehito Utsuro, Masaharu Yoshioka, Yoshihide Kawada, Noriko Kando, Tomohiro Fukuhara, Hiroshi Nakagawa, Yoji Kiyota, Noriko Kando, Proceedings of the 4th Data Engineering and Information Management Forum (DEIM Forum 2012), F5-5, Mar. 2012
Japanese, Summary national conference - Bird' Eye View Analysis on Topi Distribution of Bloggers
Kensaku Makita, Daisuke Hashimoto, Hiroko Suzuki, Takehito Utsuro, Noriko Kando, Yoshihide Kawada, Tomohiro Fukuhara, Hiroshi Nakagawa, Masaharu Yoshioka, Yoji Kiyota, Proceedings of the 4th Data Engineering and Information Management Forum (DEIM Forum 2012), F7-4, Mar. 2012
Japanese, Summary national conference - Topic Correlation and Transitions in News Blogs: With the Exmples Related to the Disasters
Daichi Koike, Daisuke Hashimoto, Kensaku Makita, Hiroko Suzuki, Takhito Utsuro, Yohihide Kawada, Masaharu Yoshioka, Noriko Kando, Tomohiro Fukuhara, Hiroshi Nakagawa, Yoji Kiyota, Yohei Seki, Proceedings of the 4th Data Engineering and Information Management Forum (DEIM Forum 2012), F1-4, Mar. 2012
Japanese, Summary national conference - Proposal of Category Description Pattern for Analyzing Wikipedia Category Structure
藤原 嵩大, 吉岡 真治, 人工知能学会全国大会論文集, 26, 1, 4, 2012
人工知能学会, Japanese - 時系列ニュース・ブログにおける話題同定に関する分析—震災を例題として—
小池大地, 牧田健作, 宇津呂武仁, 河田容英, 吉岡真治, 福原知宏, 人工知能学会全国大会論文集(CD−ROM), 2012, 0, 3E1R69, 3E1R69, 2012本論文では,震災に関する話題についての時系列のニュース記事集合,および, ブログ記事集合を対象として,トピックモデルを用いたトピック同定を行い, ニュース・ブログの間での話題の相関を分析する.本論文では,特に,各トピッ クと密接に対応するニュース記事集合・ブログ記事集合における話題がどの程度 まとまっているか,という点と,時系列バーストの有無との相関について分析を 行った結果を報告する.
, 一般社団法人 人工知能学会, Japanese - Experimental Analysis of NSContrast: Comparison and Analysis System for News Sites from Multiple Countries
吉岡 真治, 神門 典子, 関 洋平, 研究報告 情報基礎とアクセス技術(IFAT), 2011, 2, 1, 8, 26 Jul. 2011
本研究では、日中韓を中心とした複数国の新聞を比較分析する NSContrast を提案している。本システムでは、実際に毎日の新聞を新聞社のサイトから収集し、その日に注目を浴びている単語のリストや、そのような単語を含む記事数の増減や関連語などを表示することにより、従来のニュースポータルとは異なった観点からの新聞記事分析を可能にするものである。本稿では、本システムの有効性を検証するために行ったユーザ実験とその分析結果について報告する。We have already proposed a system News Site Contrast(NSContrast) that analyzing the characteristic difference between each news site and all others. The system make news article database by retrieving articles from multiple news sites in daily basis. In addition, the system have functions to analyze these articles by showing the hot keywords, related keywords for the retrieved documents. In this paper, we report the result of user experiment and analysis of the results., Japanese - Construction of Tagged Corpus for Nanodevices Development Papers
DIEB THAER M, YOSHIOKA MASAHARU, 言語処理学会年次大会発表論文集, 17th (CD-ROM), ROMBUNNO.P1-29, Mar. 2011
English - 特定トピックに関するブログ記事集合の観点分類におけるWikipediaの利用
横本 大輔, 林 東權, 牧田 健作, 宇津呂 武仁, 河田 容英, 福原 知宏, 神門 典子, 吉岡 真治, 中川 裕志, 清田 陽司, 第3回データ工学と情報マネジメントに関するフォーラム, Feb. 2011
Japanese, Introduction scientific journal - Utilizing Wikipedia as a Knowledge Source in Categorizing Topic Related Blogs into Factes
Daisuke Yokomoto, Dangkwon Lim, Kensaku Makita, Takehito Utsuro, Yasuhide Kawada, Tomohiro Fukuhara, Noriko Kando, Masaharu Yoshioka, Hiroshi Nakagawa, Yoji Kiyota, Proceedings of DEIM Forum 2011, Feb. 2011
Japanese, Introduction scientific journal - 変化検出のための極大整合連結集合
原口誠, 大久保好章, 富田悦次, 吉岡真治, 人工知能学会人工知能基本問題研究会資料, 84th, 2011 - Wikipediaを用いた地名の包含関係情報の抽出
竹中 均, 吉岡 真治, 人工知能学会全国大会論文集, 2011, 0, 2J3NFC22, 2J3NFC22, 2011東京の記述に日本という説明を加えないように、地名を含む情報を記載する場合に、その包含関係の情報を明示的に記述しないことが多い。しかし、これらの地名に関連した情報を検索する場合には、このような暗黙に仮定している関係の情報を利用する必要がある。本研究では、Wikipediaを対象にして、このような検索に有用な地名の包含関係情報の抽出方法を提案する。
, 一般社団法人 人工知能学会, Japanese - Knowledge Exploratory Project for Nano Device Design and Manufacturing : Knowledge Discovery from Experimental Records(2nd Report)Utilization of Related Metadata
吉岡真治, THAER Dieb, 冨岡克広, 原真二郎, 福井孝志, 人工知能学会全国大会論文集(CD−ROM), 25, 1, 4, 2011
人工知能学会, Japanese - Analyzing Burst of Topics in News Stream
Yusuke Takahashi, Daisuke Yokomoto, Takehito Utsuro, Masaharu Yoshioka, IPSJ SIG Notes, 6, 6, 1, 6, 2011
Among various types of recent information explosion, that in news stream is also a kind of serious problems. This paper studies issues regarding two types of modeling of information flow in news stream, namely, burst analysis and topic modeling. First, when one wants to detect a kind of topics that are paid much more attention than usual, it is usually necessary for him/her to carefully watch every article in news stream at every moment. In such a situation, it is well known in the field of time series analysis that Kleinberg's modeling of bursts is quite effective in detecting burst of key..., 一般社団法人情報処理学会, Japanese - Interactive Operation of MDS Visualization Results with Distance Metric Learning
吉岡真治, 伊藤正彦, SEBAG Michele, FEKETE Jean‐Daniel, 情報処理学会研究報告(CD-ROM), 2009, 6, ROMBUNNO.FI-98,8, 15 Apr. 2010
Japanese - Interactive Operation of MDS Visualization Results with Distance Metric Learning
YOSHIOKA MASAHARU, ITOH MASAHIKO, SEBAG MICHELE, FEKETE JEAN-DANIEL, 研究報告情報学基礎(FI), 98, 8, H1, H6, 04 Mar. 2010
多次元尺度構成法(MDS)は、高次元のデータを2次元で表現する手法である。しかし、ユーザが初期の出力結果に満足できない場合が存在する。この問題に対し、本発表では、ユーザにとって分類に役立つ次元を学習する距離尺度学習の手法を応用することによって、MDSの結果をインタラクティブに操作する方法を提案する。また、このシステムを論文の類似度を元にして作成した複数研究者の関係を可視化するシステムに適用した結果について紹介する。Multi Dimensional Scaling (MDS) is one of the popular method for visualizing multidimensional data into 2D world. However, the user may not satisfy the visualization results. In this paper, we propose a novel interactive method for modifying MDS visualization results by using the concept of distance metric learning. We also demonstrate our researcher visualization system based on this algorithm., 情報処理学会, Japanese - Wikipedia を用いた中日カタカナ翻訳辞書の作成と言語グリッドへの応用
吉岡 真治, 電子情報通信学会技術研究報告. AI, 人工知能と知識処理, 109, 424, 43, 46, 15 Feb. 2010
辞書を利用する機械翻訳システムにおいて、未登録語が多く存在する文書の翻訳結果は、質の悪いものとなる。しかし、辞書のメンテナンスは非常にコストがかかり、新語や、時事的な単語を辞書に継続的に登録することは容易ではない。本研究では、このような機械翻訳の辞書を補完するという目標のもとに、Wikipediaの言語間リンクを用いて中日の翻訳辞書を作成する方法を提案する。また、この翻訳辞書の活用事例として、多言語ニュースサイトの比較分析システムNSContrastの紹介を行う。, 一般社団法人電子情報通信学会, Japanese - ナノ知識探索プロジェクト:実験記録からの知識発見
吉岡 真治, 冨岡 克広, 原 真二郎, 福井 孝志, 人工知能学会全国大会論文集, 2010, 0, 1B33, 1B33, 2010ナノ結晶デバイスの開発プロセスにおいては、製造プロセスがデバイスの性質に大きな影響を与えるため、試行錯誤的な実験の繰り返しが必要とされる。しかし、その試行錯誤に関する知識はあまり明示化されていないため、熟練者の「匠の技」である暗黙知に依存している。本研究では、デバイス生成のための実験を行った記録を情報源として、この暗黙知の明示化を支援する実験記録データの活用システムを提案する。
, 一般社団法人 人工知能学会, Japanese - Collecting/Analyzing/Categorizing Japanese and English Blogs with Specific Topics : A Case Study
NAKASAKI HIROYUKI, ABE YUSUKE, UTSURO TAKEHITO, KAWADA YASUHIDE, FUKUHARA TOMOHIRO, KANDO NORIKO, YOSHIOKA MASAHARU, NAKAGAWA HIROSHI, KIYOTA YOJI, IPSJ SIG Notes, 194, 9, I1, I8, 16 Nov. 2009
Among other domains and topics on which some issues are frequently argued in the blogosphere, the domain of crime is one of the most seriously discussed by various kinds of bloggers. Such information on crimes in blogs is especially valuable for outsiders from abroad who are not familiar with cultures and crimes in foreign countries. This paper proposes a multilingual framework of categorizing people's concerns, reports, and experiences on crimes in their own blogs. First, we refer to Wikipedia as a multilingual terminological knowledge base, and search for Wikipedia entries describing crim..., 一般社団法人情報処理学会, Japanese - Analysis on Wikipedia Entries in Linking Topics of News and Blogs through Wikipedia
SATO YUKI, YOKOMOTO DAISUKE, NAKASAKI HIROYUKI, UTSURO TAKEHITO, YOSHIOKA MASAHARU, FUKUHARA TOMOHIRO, KANDO NORIKO, NAKAGAWA HIROSHI, KIYOTA YOJI, IPSJ SIG Notes, 194, 10, J1, J8, 16 Nov. 2009
We study complementary navigation of news and blog, where Wikipedia entries are utilized as fundamental knowledge source for linking news articles and blog feeds/posts. In the proposed framework, given a topic as the title of a Wikipedia entry, its Wikipedia entry body text is analyzed as fundamental knowledge source for the given topic, and terms strongly related to the given topic are extracted. Those terms are then used for ranking news articles and blog posts. In the scenario of complementary navigation from a news article to closely related blog posts, Japanese Wikipedia entries are ra..., 一般社団法人情報処理学会, Japanese - Editor's Message
SUMIYA K, YOSHIOKA M, SATO T, 情報処理学会論文誌. データベース, 2, 2, i, iii, 29 Jun. 2009
一般社団法人情報処理学会, Japanese - Study on Evaluation Method of NSContrast : Comparative News Article Analysis System for Characterizing the Difference among Multiple Countries
YOSHIOKA MASAHARU, IPSJ SIG Notes, 71, 1, A1, A8, 05 Jun. 2009
In this research we have already proposed a news site analysis system NSContrast that characterizes the differences among news sites from multiple countries. In this analysis system, news articles database are constructed by using following three types of news sites data; Japanese news articles written in Japanese, foreign ones written in Japanese and machine translated version of foreign ones written in foreign language. In this paper, we discuss how to evaluate the effectiveness of the NSContrast and briefly introduced the result of user experiment study., 一般社団法人情報処理学会, Japanese - Research on the Automatic Classification of Anchor Texts for utilizing Web links
OHTSUKA Hiroki, YOSHIOKA Masaharu, IPSJ SIG Notes, 94, 35, 9, 14, 18 Mar. 2009
A Link Structure among Web documents represents the relationship among these documents. Web structure analysis methods such as PageRank use this information. However, most of these methods pay little attention to the type of link (e.g., most of the methods use simple classification such as link to the same site or not). In this research, we proposed a method to classification criteria of anchor texts based on the role analysis of the text in Web documents for better link structure analysis and the users' behavior analysis on the Web page. In this paper, we propose an automatic classification method based on the result of manually classification experiment. In order to evaluate this method, we compare the result of automatic classification results with manually classified ones., Information Processing Society of Japan (IPSJ), Japanese - Wikipediaを知識源とするニュース・ブログ間のトピック対応付け
佐藤由紀, 中崎寛之, 川場真理子, 宇津呂武仁, 吉岡真治, 福原知宏, 中川裕志, 神門典子, 言語処理学会年次大会発表論文集, 15th, 300-303, 02 Mar. 2009
Japanese - 多言語ニュースの対照分析のためのWikipedia活用手法の研究
吉岡 真治, 人工知能学会全国大会論文集, 2009, 0, 2G1NFC58, 2G1NFC58, 2009本研究では、様々な国から発信される多言語ニュースに対し、機械翻訳システムを利用して対照比較を行うNSContrastの研究を行っている。本システムの分析には、翻訳間違いや表記の違いが解析に大きな影響を与えるため、正規化が必要である。本稿では、Wikipediaの言語間リンクを用いた翻訳用辞書の生成や、redirectの関係を用いた表記のバリエーションへの対応を行い、実運用における評価を行う。
, 一般社団法人 人工知能学会, Japanese - Development of Universal Abduction Studio (17th report):A Method for Creating Hypothetical Knowledge with Automatic Viewpoint Information Attachment
Sugiyama Aki, Shimomura Yoshiki, Chiba Ryosuke, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2009, 0, 211, 212, 2009
創造的設計のためには,異領域知識間の柔軟な統合が重要であると考えられており,著者らはそれを実現する統合計算機環境の開発を行っている.本稿では,知識中に記述される語彙情報から階層構造と単位次元により表現する視点情報を計算機により自動的に推定・付与するシステムを提案・開発する.また,視点情報に基づき生成する仮説的知識の創造性と有用性を可変とする操作性を,事例を用いた検証により確認する., The Japan Society for Precision Engineering - Utilization of Wikipedia Information for Comparative Analysis on Multilanguage News Sites
吉岡真治, 人工知能学会全国大会論文集(CD−ROM), 23, 1, 4, 2009
人工知能学会, Japanese - Development of Universal Abduction Studio (18th report):Implementation and verification of a web-based design knowledge acquiring method
Takahashi Kiyotaka, Sugiyama Aki, Shimomura Yoshiki, Tateyama Takeshi, Chiba Ryosuke, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2009, 0, 209, 210, 2009
前報までに、著者らは創造的設計の総合支援環境であるUniversal Abduction Studio(UAS)で利用するための設計に有用な情報を、Web上から自動取得し、当該システム用の知識ベースを構築するための基本手法を提案した。本稿では、上記提案手法を計算機上の一連のソフトウェアとして実装した内容と、実装システムにより取得された知識の設計上の利用可能性のさらに詳細な検証結果について報告する。, The Japan Society for Precision Engineering, Japanese - Structural abstraction of stories
原口 誠, 吉岡 真治, 人工知能基本問題研究会, 70, 0, 49, 52, 04 Jul. 2008
人工知能学会, Japanese - Abstract structure of stories based on lexical cohesion
井川 拓, 原口 誠, 吉岡 真治, ことば工学研究会, 28, 0, 31, 34, 28 Mar. 2008
ことば工学事務局, Japanese - Editor's Message
SUMIYA K., YOSHIOKA M., SATO T., 情報処理学会論文誌データベース(TOD), 49, 7, i, iii, 15 Mar. 2008
Japanese - A Motion Generation System with Two-Joint Muscles Installed
AKIBA RYUJI, HARAGUCHI MAKOTO, YOSHIOKA MASAHARU, 人工知能学会人工知能基本問題研究会資料, 69th, 39-42, 14 Mar. 2008
Japanese - Research on Role Analysis and Classification of Anchor Text for Utilizing Hyperlinks
OHTSUKA Hiroki, YOSHIOKA Masaharu, 全国大会講演論文集, 70, 0, 209, 210, 13 Mar. 2008
社団法人情報処理学会, Japanese - A knowledge representation for the creative design supports using knowledge tagged by the surface case
杉山明紀, 下村芳樹, 吉岡真治, 武田英明, 日本機械学会設計工学・システム部門講演会論文集(CD-ROM), 18th, 2008 - IR interface for contrasting multiple news sites
Masaharu Yoshioka, INFORMATION RETRIEVAL TECHNOLOGY, 4993, 508, 513, 2008, [Peer-reviewed]
In order to utilize news articles from multiple news sites, it is better to understand the characteristics of each news site. In this paper, a concept of contrast set mining is applied for analyzing the characteristic difference between each news site and all others. The News Site Contrast (NSContrast) system is also proposed based on this mining technique. This system is applied to a news article database constructed from multiple news sites from different countries in order to evaluate its effectiveness., SPRINGER-VERLAG BERLIN, English - Web文書におけるアンカーテキストの役割分析と自動分類の研究
大塚 博紀, 吉岡 真治, 人工知能学会全国大会論文集, 8, 0, 203, 203, 2008
本研究では、Webのリンク構造を分類して利用することを目標として、アンカーテキストの情報を用いたリンクの役割分析を行っている。本発表では、役割分類の基準とそれに基づいた自動分類の手法について提案する。, 一般社団法人 人工知能学会, Japanese - 創造的設計支援のための格文法と物理量情報を用いるタグ付け知識作成手法
杉山明紀, 下村芳樹, 千葉龍介, 吉岡真治, 武田英明, Designシンポジウム講演論文集, 2008, 441-446, 2008
Japanese - Research on Role Analysis and Automatic Classification of Anchor Texts in Web Documents
OTSUKA HIROKI, YOSHIOKA MASAHARU, 人工知能学会全国大会論文集(CD−ROM), 22, 1, 3, 2008
人工知能学会, Japanese - Development of Universal Abduction Studio(13th Report)
Nakanishi Masatoshi, Suzuki Ryo, Simomura Yoshiki, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2008, 0, 435, 436, 2008
本研究では創造的設計を総合的に支援する計算機環境の構築を目指している。著者らは、これまでに創造的設計を支援するための知識記述形式や推論機構、オントロジー利用の枠組みなどを提案してきた。しかしながらこれらを用いて創造的な設計を支援するためには以上の構成要素群を統合する必要がある。本稿では創造的設計を支援するための統合計算機環境を提案し、その具体的な実装構造を例示する。, The Japan Society for Precision Engineering, Japanese - Development of Universal Abduction Studio (14th report):Web-based expansion scheme for a design knowledge information database
Takahashi Kiyotaka, Suzuki Ryou, Shimomura Yoshiki, Tateyama Takeshi, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2008, 0, 437, 438, 2008
In recent years, comprehensive problem solving by a designer has been required as the demand for an artifact grows and complicates more. In the above background, we have proposed Universal Abduction Studio (UAS), a computer environment, which supports a creative design synthetically. However, it is difficult for a designer to acquire the multiple domain knowledge on the grounds that the knowledge is accumulated manually. This paper proposes a method of expanding the design knowledge database used in UAS by extracting useful information for design from the resource on the web. Also, the authors propose a filtering method for reducing the information which isn't useful for design from the database. At the end, the effectively of the methods is verified., The Japan Society for Precision Engineering, Japanese - Development of Universal Abduction Studio (15th report):A knowledge representation format to support creative design– a method for knowledge tagging based on physical quantity
Sugiyama Aki, Shimomura Yoshiki, Chiba Ryosuke, Takeda Hideaki, Yoshioka Masaharu, Proceedings of JSPE Semestrial Meeting, 2008, 0, 647, 648, 2008
著者らは,創造的設計を支援する統合計算機環境の構築を行っている.これまでに,語彙情報をタグ付けした知識から概念類似度を基に生成される仮説的知識の有用性向上手法を提案しているが,場面に即した仮説的知識を効率的に取得するためには語彙の属性に基づくタグ付け手法とそれを用いた推論が有効である.本稿では,知識に対して語彙情報に基づく物理量をタグ付けする知識表現手法を提案し,その有効性を検証する., The Japan Society for Precision Engineering, Japanese - Development of Universal Abduction Studio (16th Report):Web-based knowledge database extension method for supporting design
Takahashi Kiyotaka, Sugiyama Aki, Shimomura Yoshiki, Tateyama Takeshi, Chiba Ryosuke, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2008, 0, 649, 650, 2008
近年,人工物に対する要求の高度化・複雑化に伴い,設計者に対する負担が一層増大している.以上の背景のもと,著者らは創造的設計を総合的に支援する計算機環境Universal Abduction Studio(UAS)を提案している.本稿では,設計に有用な情報をWeb情報より自動的に取得,UASで利用可能な知識表現に変換することにより,UASの設計知識データベースを効率的に拡充するための手法を提案する., The Japan Society for Precision Engineering, Japanese - 1311 A knowledge representation for the creative design support using knowledge tagged by the surface case
SUGIYAMA Aki, SHIMOMURA Yoshiki, YOSHIOKA Masaharu, TAKEDA Hideaki, The Proceedings of Design & Systems Conference, 2008, 0, 181, 185, 2008
In order to realize a creative design, it is thought to be important to unify the knowledge from different knowledge areas flexibly. The computer environment that supports designers in an integrated way has been developed. The extended method of the knowledge representation forms for increasing the usefulness of the hypothetical knowledge from the knowledge tagged with vocabulary information was proposed. However, the current method isn't always capable of obtaining the useful knowledge, since it doesn't consider the context. In order to acquire the hypothetical knowledge adapted to the context efficiently, the tagging technique based on the attribute of a vocabulary and the reasoning using it are effective. In this paper, the authors propose knowledge representation forms for tagging physical quantity based on vocabulary describing knowledge, and verify the usefulness of the proposed method., The Japan Society of Mechanical Engineers, Japanese - 2302 Web-based knowledge information database extension method for supporting design
TAKAHASHI Kiyotaka, SHIMOMURA Yoshiki, TATEYAMA Takeshi, YOSHIOKA Masaharu, TAKEDA Hideaki, The Proceedings of Design & Systems Conference, 2008, 0, 305, 309, 2008
In recent years, comprehensive problem solving by a designer has been required as the demand for an artifact grows and complicates more. In the above background, we have proposed Universal Abduction Studio (UAS), a computer environment, which supports a creative design synthetically. However, it is difficult for designers to acquire knowledge in multiple domains efficiently because they have to accumulate the design knowledge by the handwork. Thus far, we proposed Web-based design knowledge information database extension method. In this paper, we propose the conversion method acquired design knowledge information from the resource on Web into knowledge expression for UAS., The Japan Society of Mechanical Engineers, Japanese - Editor's Message
SUMIYA K., YOSHIOKA M., SATO T., 情報処理学会論文誌データベース(TOD), 48, 20, i, iii, 15 Dec. 2007
Japanese - Information Access Technology for Utilizing Wide Variety of Text Information
吉岡 真治, Human interface, 9, 4, 269, 274, 25 Nov. 2007
ヒュ-マンインタフェ-ス学会, Japanese - A knowledge representation method based on case grammar classification and knowledge templates
SUGIYAMA AKI, SUZUKI RYO, NAKANISHI MASATOSHI, SHIMOMURA YOSHIKI, YOSHIOKA MASAHARU, TAKEDA HIDEAKI, 日本機械学会設計工学・システム部門講演会講演論文集, 17th, 286-287, 30 Oct. 2007
Japanese - Univeral Abduction Studioの開発(第12報)—格分類と知識テンプレートに基づく知識表現—
杉山明紀, 鈴木遼, 中西雅俊, 下村芳樹, 吉岡真治, 武田英明, 精密工学会大会学術講演会講演論文集, 2007, G67, 03 Sep. 2007
Japanese - 1001 SIG Nights : Special Interest Group on Fundamental Informatics
YOSHIOKA Masaharu, IPSJ Magazine, 48, 8, 920, 921, 15 Aug. 2007
Information Processing Society of Japan (IPSJ), Japanese - 適合的汎化に基づく情報検索システムの実験的研究
吉岡 真治, 原口誠, NICT-EDR電子化辞書シンポジウム, -, -, 67, 76, Apr. 2007
Japanese, Introduction scientific journal - 複数のニュース源の差異を考慮したニュース分析の研究,「大規模Web研究基盤上での自然言語処理・情報検索」
吉岡真治, 言語処理学会第13回年次大会ワークショップ (W2), 27, 30, Mar. 2007
Japanese, Introduction scientific journal - Editors' Introduction to "Grant in Aid for Scientific Research on Priority Areas : Cyber Infrastructure for the Information Explosion Era"(
Grant in Aid for Scientific Research on Priority Areas: Cyber Infrastructure for the Information Explosion Era)
Masaharu Yoshioka, Ikuko Yairi, Journal of Japanese Society for Artificial Intelligence, 22, 2, 208, 208, 01 Mar. 2007
社団法人人工知能学会, Japanese - Editors' Introduction to "Grant in Aid for Scientific Research on Priority Areas : Cyber Infrastructure for the Information Explosion Era"
YOSHIOKA Masaharu, YAIRI Ikuko, 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 22, 2, 208, 208, 01 Mar. 2007
Japanese - Editors' Introduction to "Grant in Aid for Scientific Research on Priority Areas : Cyber Infrastructure for the Information Explosion Era"(
Grant in Aid for Scientific Research on Priority Areas: Cyber Infrastructure for the Information Explosion Era)
YOSHIOKA Masaharu, YAIRI Ikuko, Journal of the Japanese Society for Artificial Intelligence, 22, 2, 208, 208, 2007
The Japanese Society for Artificial Intelligence, Japanese - 情報源の特性を考慮した情報アクセス(<特集>編集委員2007年の抱負)
吉岡 真治, Masaharu Yoshioka, 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 22, 1, 15, 15, 01 Jan. 2007
社団法人人工知能学会, Japanese - 3110 A knowledge representation method based on case grammar classification and knowledge templates
SUGIYAMA Aki, SUZUKI Ryo, NAKANISHI Masatoshi, SHIMOMURA Yoshiki, YOSHIOKA Masaharu, TAKEDA Hideaki, The Proceedings of Design & Systems Conference, 2007, 0, 286, 287, 2007
In order to realize a creative design, it is thought to be important to unify the different knowledge group flexibly, and the concept of the computer environment which supports designers in an integrated way has been proposed. Our project, called Universal Abduction Studio, aims at assisting designers in obtaining design solutions through the dynamic interaction between a system and a designer. In this report, we propose the extended method of the knowledge representation forms for increasing the usefulness of hypothetical knowledge, after considering the existing reasoning environment. Finally, the usefulness of the proposed method is verified., The Japan Society of Mechanical Engineers, Japanese - Development of Universal Abduction Studio (12th report)
Sugiyama Aki, Suzuki Ryo, Nakanisi Masatosi, Shimomura Yoshiki, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2007, 0, 523, 524, 2007
創造的な設計を実現するためには異領域知識間の柔軟な統合が重要と考えられ,それを総合的に支援する計算機環境であるUniversal Abduction Studioの概念が提案されている.本稿では,これまでに著者らが提案した推論機構に関する考察を行なった上で,計算機が導出する仮説的知識の有用性を,各分類と知識テンプレートの導入により向上させる知識表現形式の拡張方法を提案し,その有効性を検証する., The Japan Society for Precision Engineering - Web-based expansion scheme for a design knowledge information database
Takahashi Kiyotaka, Suzuki Ryou, Shimomura Yoshiki, Tateyama Takeshi, Yoshioka Masaharu, Takeda Hideaki, Proceedings of the Conference of Transdisciplinary Federation of Science and Technology, 2007, 0, 45, 45, 2007
In recent years, comprehensive problem solving by a designer has been required as the demand to an artifact develops and complicates more. In the above background, we have proposed Universal Abduction Studio (UAS), a computer environment that supports a creative design synthetically. This paper proposes the method of expanding the design knowledge database used in UAS by extracting useful information for design from the resource on the web., Transdisciplinary Federation of Science and Technology, Japanese - 特許検索のための物理法則概念による文書分類の研究
吉岡真治, 落合翔平, 精密工学会大会学術講演会講演論文集, 2007, 0, 513, 514, 2007
TRIZなどに代表されるようにアイディア検索のために特許を利用する方法が提案されている。このような状況では、表面上の表記に注目した同分野の特許を検索するのではなく、類似の問題を扱っている特許を検索することが重要となる。本研究では、アイディアの根幹にある物理法則をキーとした検索を考え、学習により文書に対して関連物理法則のラベルをつけ、分類する方法を提案する。, 公益社団法人 精密工学会, Japanese - Classification of Anchor Text for Web Information Application
Yoshioka Masaharu, IPSJ SIG Notes, 2006, 94, 27, 33, 12 Sep. 2006
Most significant characteristics of the Web information is description about a relationship among different pages by using hyperlinks and anchor texts. In this research, anchor texts extracts from large web test collection NTCIR nw100g are investigated for a proposal of a functional classification. A strategy for automatic classification of anchor texts in html pages and systems that utilize this classification are also proposed in this paper., Information Processing Society of Japan (IPSJ), Japanese - Classification of Anchor Text for Web Information Application
YOSHIOKA Masaharu, IPSJ SIG Notes, 84, 94, 27, 33, 12 Sep. 2006
Most significant characteristics of the Web information is description about a relationship among different pages by using hyperlinks and anchor texts. In this research, anchor texts extracts from large web test collection NTCIR nw100g are investigated for a proposal of a functional classification. A strategy for automatic classification of anchor texts in html pages and systems that utilize this classification are also proposed in this paper., Information Processing Society of Japan (IPSJ), Japanese - Universal Abduction Studio Hypothetical Knowledge Generation Support Environment for Creative Design Universal Abductioni Studio
吉岡真治, 佐藤孝彦, 森本憲悟, 武田英明, 下村芳樹, 人工知能学会人工知能基本問題研究会資料, 62, 0, 71, 76, 27 Mar. 2006
人工知能学会, Japanese - Universal Abduction Studioの開発(第9報)—創造的設計支援のための知識記述エディタの開発—
伊東真樹, 森本憲悟, 下村芳樹, 武田英明, 吉岡真治, 精密工学会大会学術講演会講演論文集, 2006, M22, 01 Mar. 2006
Japanese - Universal Abduction Studioの開発(第8報)—創造的設計支援のためのオントロジー構造—
中西雅俊, 森本憲悟, 下村芳樹, 武田英明, 吉岡真治, 精密工学会大会学術講演会講演論文集, 2006, M21, 01 Mar. 2006
Japanese - Universal Abduction Studioの開発(第10報)—対話型進化的計算法に基づく仮説的な設計知識の生成手法の提案—
森本憲悟, 下村芳樹, 吉岡真治, 武田英明, 上田完次, 精密工学会大会学術講演会講演論文集, 2006, M23, 01 Mar. 2006
Japanese - 115 Hypothetical Knowledge Generation Method for Creative Design
YOSHIOKA Masaharu, SATOH Takahiko, SUGIYAMA Aki, MORIMOTO Kengo, TAKEDA Hideaki, SHIMOMURA Yoshiki, Designシンポジウム講演論文集, 2006, 0, 95, 100, 2006
In order to design new artifacts that are different from existing ones, it is crucial to have new knowledge for the design problem. We have already confirmed that combination of existing knowledge from different domain may useful for this process by analyzing design cases. In this paper, we briefly review our hypothetical knowledge generation support environment "Universal Abduction Studio" (UAS) and propose an automatic method to organize design document for this environment., 社団法人日本機械学会, Japanese - 2303 A Method for terminology management based on transparent ontology
NAKANISHI Masatoshi, SUZUKI Ryo, SHIMOMURA Yoshiki, TAKEDA Hideaki, YOSHIOKA Masaharu, The Proceedings of Design & Systems Conference, 2006, 0, 206, 207, 2006
We argue that abduction for integrating theories can be a basic principle to formalize creative design processes. Based on this principle, we have proposed Universal Abduction Studio, a design environment in which designers combine different theories to arrive at better design. In this new approach to computational support of conceptual design, the system should offer various types of abductive reasoning from which designers can select an interesting design method. This paper aims at proposing the method of handling terminology by general common ontology and managing in the multiple. We also discuss technologies to implement proposed method of managing terminology., The Japan Society of Mechanical Engineers, Japanese - 創造的設計支援に用いる仮説的設計知識生成のための自動文書タグ付け手法の提案
吉岡 真治, 佐藤 孝彦, 森本 憲悟, 武田 英明, 下村 芳樹, 人工知能学会全国大会論文集, 6, 0, 202, 202, 2006
従来の設計解とは本質的に異なる創造的な設計解を作るためには、既存の知識の単純な組合わせではなく、新たな設計知識の利用が不可欠である。本研究では、これまで、異なる領域知識間に存在する類似性をキーとしたアナロジーを用いて、仮説的設計知識の生成手法を提案してきた。本発表では、このアナロジーを行なうために必要な文書情報のタグ付けを自動化するための方法論について述べる。, 一般社団法人 人工知能学会, Japanese - Development of Universal Abduction Studio(9th report)
Itou Masaki, Morimoto Kengo, Shimomura Yoshiki, Takeda Hideaki, Yoshioka Masaharu, Proceedings of JSPE Semestrial Meeting, 2006, 0, 1001, 1002, 2006
設計課題に関連する領域知識を用いるだけでは,創造的な設計解を得ることは困難である.本研究では創造的設計実現のためには,設計課題に関する領域知識とそれとは異なる領域知識の統合が重要であると考え,これを総合的に支援する計算機環境の構築を目指している.本稿ではより高度な知識統合を実現するために,知識の意味的構造に基づく統合を可能とする知識表現形式とその形式に従い知識を記述するためのエディタを提案する., The Japan Society for Precision Engineering - Development of Universal Abduction Studio (10th report)
Morimoto Kengo, Shimomura Yoshiki, Takeda Hideaki, Yoshioka Masaharu, Ueda Kanji, Proceedings of JSPE Semestrial Meeting, 2006, 0, 1003, 1004, 2006
It is important to unify different knowledge groups, which exist independently, and to extend knowledge gradually in order to obtain creative design solutions that are not obtained only with the existing design knowledge. With the aim of development of ″Universal Abduction Studio″, which is a system supporting creative design, we propose a method for creating hypotheses through interactions between designer and machine. We develop a prototype system and demonstrate that the system can create new ideas., The Japan Society for Precision Engineering - Proposal of Automatic Document Tagging Method for Hypothetical Knowledge Generation in Creative Design Support System
YOSHIOKA MASAHARU, SATO TAKAHIKO, MORIMOTO KENGO, TAKEDA HIDEAKI, SHIMOMURA YOSHIKI, 人工知能学会全国大会論文集(CD−ROM), 20, 1, 4, 2006
人工知能学会, Japanese - Development of Universal Abduction Studio(11th Report)
Nakanishi Masatoshi, Suzuki Ryo, Shimomura Yoshiki, Takeda Hideaki, Yoshioka Masaharu, Proceedings of JSPE Semestrial Meeting, 2006, 0, 109, 110, 2006
We argue that abduction for integrating theories can be a basic principle to formalize creative design processes. Based on this principle, we have proposed Universal Abduction Studio, a design environment in which designers combine different theories to arrive at better design. In this new approach to computational support of conceptual design, the system should offer various types of abductive reasoning from which designers can select an interesting design method. This paper aims at proposing the method of handling terminology by general common ontology and managing in the multiple. We also discuss technologies to implement proposed method of managing terminology., The Japan Society for Precision Engineering, Japanese - Overview of the NTCIR-5 WEB Query Term Expansion Subtask. In Proceedings of the Fifth NTCIR Workshop Meeting on Evaluation of Information Access Technologies, National Institute of Informatics, 2005.
Masaharu Yoshioka, Information Retrieval, Question Answering and Cross-Lingual Information Access, 443, 454, Dec. 2005
English, Introduction scientific journal - Proposal of an Analysis Method for Evaluating the Effectivess of Query Term Expansion in Information Retrievval
YOSHIOKA Masaharu, IPSJ SIG Notes, 81, 108, 53, 60, 02 Nov. 2005
Query term expansion is a technique to modify initial query by adding some terms that represents users' information need for achieving higher retrieval performance. However, a good framework for evaluating this technique in isolation does not exist because the effect of query expansion depends on several issues. In this paper, I enumerate several features that may affect the quality of this technique and analyze retrieval experiment results by using these features., Information Processing Society of Japan (IPSJ), Japanese - Proposal of an analysis method for evaluating the effectives of query term expansion in information retrieval
Yoshioka Masaharu, Digital libraries, 29, 55, 62, Nov. 2005
ディジタル図書館編集委員会, Japanese - Analysis on Effectiveness of Query Terms in Information Retrieval Test Collection
YOSHIOKA Masaharu, IPSJ SIG Notes, 136, 42, 25, 30, 19 Sep. 2005
In many cases, it is not easy for a user of Information Retrieval (IR) to select appropriate query terms for representing his/her information need. Since this mismatch between query terms and information need affects the topic difficulty, it is better to have a framework to evaluate the effectiveness of the query terms. Therefore, in this paper, I propose some feature quantities that characterize quality of initial query in test collection. I also analyze correlation coefficient between these quantities and retrieval performance by using NTCIR-4 web test collection., Information Processing Society of Japan (IPSJ), Japanese - Double-Decker RFID Tags : Basic Structures and Applications(Paper Session 8 (uT Technology))
Shinichi Minato, Masaharu Yoshioka, 情報処理学会研究報告. UBI, [ユビキタスコンピューティングシステム], 2005, 60, 497, 501, 09 Jun. 2005
For tracing or finding the lost goods, we need active-type of RFID tags with built-in batteries, but those active tags have significant drawbacks such as the higher hardware cost and the finite battery lifetime. In this paper, we propose "double-decker" RFID tags, which can be separated into the two parts of ID tags. The new type of RFID tag is useful and cost-effective for tracing/keeping of a large number of goods. We present the basic structures of the double-decker tags and show some typical applications., Information Processing Society of Japan (IPSJ), English - Analysis on Effectiveness of Query Terms in Information Retrieval Test Collection
Yoshioka Masaharu, IPSJ SIG Notes, 2005, 42, 25, 30, 19 May 2005
In many cases, it is not easy for a user of Information Retrieval (IR) to select appropriate query terms for representing his/her information need. Since this mismatch between query terms and information need affects the topic difficulty, it is better to have a framework to evaluate the effectiveness of the query terms. Therefore, in this paper, I propose some feature quantities that characterize quality of initial query in test collection. I also analyze correlation coefficient between these quantities and retrieval performance by using NTCIR-4 web test collection., 一般社団法人情報処理学会, Japanese - 人間と人工知能の得手不得手(<特集>ようこそ人工知能の世界へ : 編集委員今年の初夢)
吉岡 真治, Masaharu Yoshioka, 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 20, 3, 268, 268, 01 May 2005
社団法人人工知能学会, Japanese - 人間と人工知能の得手不得手
吉岡 真治, 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 20, 3, 268, 268, 01 May 2005
Japanese - 人間と人工知能の得手不得手(<特集>ようこそ人工知能の世界へ : 編集委員今年の初夢)
吉岡 真治, 人工知能, 20, 3, 268, 268, 2005
一般社団法人 人工知能学会, Japanese - 1405 A Method for Generating Hypothetic Design Knowledge with Analogy
MORIMOTO Kengo, ONIKI Wataru, SHIMOMURA Yoshiki, YOSHIOKA Masaharu, TAKEDA Hideaki, UEDA Kanji, The Proceedings of Design & Systems Conference, 2005, 0, 168, 171, 2005
It is important to unify different knowledge groups, which exist independently, and to extend knowledge gradually in order to obtain creative design solutions that are not obtained only with the existing design knowledge. With the aim of development of "Universal Abduction Studio", which is a system supporting creative design, a knowledge representation and a matching algorithm for creating hypothetic knowledge is proposed in this paper., The Japan Society of Mechanical Engineers, Japanese - 2306 Research on Extraction of Stake Holder Network for Program Management
YOSHIOKA Masaharu, The Proceedings of Design & Systems Conference, 2005, 0, 334, 337, 2005
Program management aims to support a management for a large and complex project. We have already proposed a framework for supporting this management process by reviewing a program from three different aspects (i.e., process planning, related information, and stake holder networks). However, this system only supports to represent and review a program by using manually inputted information. In this paper, I proposed a system that extract stakeholders information from a text to support constructing stake holder network. In this system, we categorize stakeholders in two types. One is collaborator and the other is opponent. For extracting these two types of stakeholders, we collected typical expression for describing them and made pattern match rule for extraction. For evaluating our methodology, we implemented the prototype system that uses Japanese newspaper articles as a resource for extraction., The Japan Society of Mechanical Engineers, Japanese - 2827 A Design Support Environment for Creative Thinking Based on a Tagging Method
ONIKI Wataru, MORIMOTO Kengo, SHIMOMURA Yoshiki, YOSHIOKA Masaharu, TAKEDA Hideaki, The proceedings of the JSME annual meeting, 2005, 0, 249, 250, 2005
When a designer confronts a difficult design problem which cannot be solved only with his/her own knowledge, it is necessary to amplify it by introducing new perspectives and ideas for design. The authors have been developing a integrated computer aided design system for creative design. In this report, the authors propose a knowledge description framework which enables computers to analogize a bit of knowledge to another in a different field. Additionally, a tagging method of knowledge by RDF/XML and the implement of an editing tool for the method are mentioned in this report., The Japan Society of Mechanical Engineers, Japanese - プログラムマネジメント支援のためのテキストからの利害関係者の抽出と分類
吉田 守泰, 吉岡 真治, 人工知能学会全国大会論文集, 5, 0, 95, 95, 2005
大規模で複雑な事業などの支援を目的としたプログラムマネジメントにおいて、関係する利害関係者を理解することは重要な意味を持つ。本研究では、この情報を、新聞記事などのテキストからパタンを用いて抽出・分類する方法を提案する。, 一般社団法人 人工知能学会, Japanese - A Method for Generating Hypothetic Design Knowledge with Analogy
MORIMOTO KENGO, ONIKI WATARU, FUJIMOTO YUTAKA, SHIMOMURA YOSHIKI, YOSHIOKA MASAHARU, TAKEDA HIDEAKI, 人工知能学会全国大会論文集(CD−ROM), 5, 0, 282, 282, 2005
本稿では,設計における問題解決に有効な仮説的知識を生成するために,概念ネッワークから知識モデルを生成し,そのモデルに対してアナロジーを適用する手法を提案する.さらに実験により本手法の有効性を検証する., 一般社団法人 人工知能学会, Japanese - Extraction and Classification of Stakeholder from Text to Support Program Management
YOSHIDA MORIYASU, YOSHIOKA MASAHARU, 人工知能学会全国大会論文集(CD−ROM), 19, 1, 4, 2005
人工知能学会, Japanese - Development of Universal Abduction Studio (7th Report)
Morimoto Kengo, Oniki Wataru, Fujimoto Yutaka, Shimomura Yoshiki, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2005, 0, 183, 183, 2005
It is important to unify different knowledge groups, which exist independently, and to extend knowledge gradually in order to obtain creative design solutions that are not obtained only with the existing design knowledge. Authors have proposed some knowledge representations for realizing knowledge extension gradual with the aim of development of “Universal Abduction Studio”, which is a system supporting creative design. In this paper, matching algorithm based on construction with deep case and validation checking algorithm with context are proposed., The Japan Society for Precision Engineering, Japanese - Development of Universal Abduction Studio (6th Report)
Oniki Wataru, Morimoto Kengo, Fujimoto Yutaka, Shimomura Yoshiki, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2005, 0, 182, 182, 2005
When a designer confronts a difficult design problem which cannot be solved only with his/her own knowledge, it is necessary to amplify it by introducing new perspectives and ideas for design. The authors have been developing a integrated computer aided design system for creative design, and have proposed “knowledge scheme” to manage knowledge operations based on the meaning of knowledge in the last report. In this report, the authors propose a knowledge description format based on the knowledge scheme in the form of RDF/XML. In addition, the implementation of a knowledge editor to support the knowledge description is mentioned., The Japan Society for Precision Engineering, Japanese - Universal Abduction Studioの開発(第5報)—UASのための知識スキーム—
森本憲悟, 下村芳樹, 藤本裕, 武田英明, 吉岡真治, 精密工学会大会学術講演会講演論文集, 2004, A68, 01 Sep. 2004
Japanese - Research on Documentation Management System for Problem Solving based on Task Similarity of a Problem
吉岡真治, 五十嵐優, 人工知能学会人工知能基本問題研究会資料, 56, 0, 91, 96, 25 Jul. 2004
人工知能学会, Japanese - A Query-Driven Construction of Maximal Analogies from Documents
HARAGUCHI Makoto, YOSHIOKA Masaharu, IPSJ SIG Notes, 162, 73, 7, 10, 15 Jul. 2004
We present here a method for constructing a common abstract event sequence from two or more documents also regarded as event sequences. To reduce the computational cost and to improve the precision of relevance judgement made by the abstract event sequences, we introduce a notion of base queries and consider a retrieval process to find an instance of base query that is similar to a given family of instance documents. By the base query and its instance documents, we can design a divide and conquer strategy for the construction of abstract common event sequence from the instances., Information Processing Society of Japan (IPSJ), Japanese - Development of a Design Methodology for Upgradability involving Changes of Functions
KONDO SHINSUKE, ISHIGAMI YOSHIAKI, UMEDA YASUSHI, SHIMOMURA YOSHIKI, YOSHIOKA MASAHARU, Designシンポジウム講演論文集, 2004, 227-230, 08 Jul. 2004
Japanese - Proposal of the Universal Abduction Studio
FUJIMOTO YUTAKA, SHIMOMURA YOSHIKI, TAKEDA HIDEAKI, YOSHIOKA MASAHARU, Designシンポジウム講演論文集, 2004, 99-104, 08 Jul. 2004
Japanese - Research on Information Retrieval System based on Adaptive Generalization (2nd Report) : Construction of IR User Interface that focuses on Generalized Concept with High Coverage
YOSHIOKA Masaharu, HARAGUCHI Makoto, IPSJ SIG Notes, 74, 36, 87, 93, 26 Mar. 2004
It is not easy for a user of Information Retrieval (IR) system to select appropriate keywords. In this research, we proposed a new IR system that uses adaptive generalization of keywords. When the system can selects appropriate generalization by estimating user's intent, the system can generate good keywords that have high readability and good retrieval performance. In this report, we proposed new IR user interface that displaies generalized concepts by using relevant documents information. In addition, we propose a method to construct Boolean retrieval formula by using these generalized concepts and apply this method for Web information retrieval., Information Processing Society of Japan (IPSJ), Japanese - Research on Information Retrieval System based on Adaptive Generalization (2nd Report) -Construction of IR User Interface that focuses on Generalized Concept with High Coverage-
吉岡 真治, 原口 誠, 情報処理学会研究報告デジタルドキュメント(DD), 2004, 36, 87, 93, 26 Mar. 2004
情報検索システムを利用する検索者にとって、適切なキーワードを選択することは必ずしも容易なことではない。本研究では、検索者にも理解しやすい概念階層の汎化という考え方を利用して、ユーザの検索意図を明示化すると共に、精度落ちを抑えた情報検索システムを提案している。本報では、汎化概念の関連文書における網羅性に注目し、検索に役に立つと考えられる汎化概念を明示化するインタフェースの作成について報告する。また、この網羅性に注目することにより、検索意図をより明確に表現するBoolean検索式の構築の支援とWeb検索への応用についても述べる。It is not easy for a user of Information Retrieval (IR) system to select appropriate keywords. In this research, we proposed a new IR system that uses adaptive generalization of keywords. When the system can selects appropriate generalization by estimating user's intent, the system can generate good keywords that have high readability and good retrieval performance. In this report, we proposed new IR user interface that displaies generalized concepts by using relevant documents information. In addition, we propose a method to construct Boolean retrieval formula by using these generalized concepts and apply this method for Web information retrieval., Japanese - Study on the combination of probabilistic and boolean ir models for www documents retrieval
YOSHIOKA Masaharu, Working Notes of the Fourth NTCIR Workshop Meeting, 2004, 9, 16, 2004 - A Tactile Sensor Based on Computer Tomograph
MORIMOTO Kengo, FUJIMOTO Yutaka, SHIMOMURA Yoshiki, TAKEDA Hideaki, YOSHIOKA Masaharu, The Proceedings of Design & Systems Conference, 2004, 0, 332, 335, 2004
In order to obtain a creative design solution, which is not obtained only with existing design knowledge, it is required to unify different knowledge groups, which exist independently, and to perform gradual extension of knowledge. Authors have proposed some knowledge representations and algorithms of knowledge integration for realizing knowledge extension with the aim of development of "Universal Abduction Studio", which supports creative design synthetically. In this paper, the fundamental framework for dealing with information as knowledge is proposed. And the representation and the knowledge integration based on it are proposed., The Japan Society of Mechanical Engineers, Japanese - 設計情報の知的処理のためのタグ付け方法の検討
武田 英明, 藤本 裕, 吉岡 真治, 下村 芳樹, 人工知能学会全国大会論文集, 4, 0, 287, 287, 2004
設計情報を計算機によって知的処理させるためには,テキストに機械処理可能なタグ付けをする必要がある.設計の場合,多様な文脈があるため,一意のタグづけは難しい.そこで本研究では,複数の解釈が可能なタグ付け方法を提案し,書籍の文章を例題にタグづけを試験的におこなった., 一般社団法人 人工知能学会, Japanese - Development of Universal Abduction Studio (5th Report)
Morimoto Kengo, Shimomura Yoshiki, Fujimoto Yutaka, Takeda Hideaki, Yoshioka Masaharu, Proceedings of JSPE Semestrial Meeting, 2004, 0, 45, 45, 2004
既存の設計知識だけでは得られないような創造的な設計解を得るためには、独立<br>に存在する異なる知識群を統合し、知識の段階的拡張を行うことが必要である。<br>著者らは創造的設計を総合的に支援する環境・UASの構築を目指し、これまでに<br>段階的な知識拡張を実現するための、複数の知識表現形式や知識統合の手法を提<br>案してきた。本報では自然言語などによって記述された情報を知識として取扱う<br>ための基本的枠組みを提案する。, The Japan Society for Precision Engineering - 問題解決のための知識マネージメントシステムの研究(第3報) -概念階層を利用した類似事例の検索-
吉岡 真治, 五十嵐 優, 人工知能学会全国大会論文集, 4, 0, 130, 130, 2004
本研究では、ユーザが持つ問題に対し、類似事例などを提示しながら、問題解決を支援する知識マネージメントシステムの研究を行っている。本報では、オントロジーに基づく概念定義を利用した類似事例の検索と再利用の方法について述べる。, 一般社団法人 人工知能学会, Japanese - Tagging for intelligent processing of design information
TAKEDA HIDEAKI, FUJIMOTO YUTAKA, YOSHIOKA MASAHARU, SHIMOMURA YOSHIKI, 人工知能学会全国大会論文集(CD−ROM), 18, 1, 3, 2004
人工知能学会, Japanese - Development of Support System for Upgradability involving Changes of Functions
石上佳照, 近藤伸亮, 梅田靖, 下村芳樹, 吉岡真治, Proceedings of JSPE Semestrial Meeting, 2004, 0, 297, 297, 2004
アップグレード設計は、単に製品の物理的寿命を延長するだけでなく、機能的寿命を延長する設計手法である。本稿では、機能変更を伴う(e.g. 片面コピー機に二世代目で両面コピー機能を追加)アップグレード設計を扱う。従来、著者らが提案してきたアップグレード設計方法論に基づき、設計支援システムを開発し、実行例を述べ、本方法論の有効性を検証する。, The Japan Society for Precision Engineering, Japanese - Development of Universal Abduction Studio (4th report)
Fujimoto Yutaka, Nomaguchi Yutaka, Simomura Yoshiki, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2004, 0, 197, 197, 2004
対象固有の領域知識を用いるだけでは、既存の設計解を改良することは可能であっても創造性の高い設計を支援することは困難である。即ち、創造性の高い設計を支援するためには、異なる領域に属する知識を互いに関連づけることによる問題の再定義や、新しい知識を作成するための手段が必要である。本報では、前報ですでに提案したオントロジーを用いた知識表現の枠組みとその評価結果に基づき、より柔軟な知識利用を実現するための新しい知識表現の手法を提案するとともに、その有効性について検証する。, The Japan Society for Precision Engineering, Japanese - NTCIRプロジェクト
吉岡 真治, 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 19, 1, 127, 128, 01 Jan. 2004
社団法人人工知能学会, Japanese - 言語情報処理ポータル
吉岡 真治, 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 18, 3, 367, 368, 01 May 2003
社団法人人工知能学会, Japanese - Editor's Introduction to "NTCIR from the View Point of Participant"(
Intelligent Manufacturing Systems:IMS)
YOSHIOKA Masaharu, Journal of Japanese Society for Artificial Intelligence, 18, 2, 107, 107, 01 Mar. 2003
社団法人人工知能学会, Japanese - Editor's Introduction to "Intelligent Manufacturing Systems:IMS"(
Intelligent Manufacturing Systems:IMS)
Masaharu Yoshioka, Journal of the Japanese Society for Artificial Intelligence, 18, 2, 107, 107, 2003
The Japanese Society for Artificial Intelligence, Japanese - Development of Design for Upgradability Considering Future Uncertainty
ISHIGAMI Yoshiaki, KONDOH Shinsuke, UMEDA Yasushi, SHIMOMURA Yoshiki, YOSHIOKA Masaharu, The proceedings of the JSME annual meeting, 2003, 0, 323, 324, 2003
The objective of upgrade design is to extend product life so as to reduce consumption volume of resources and energy usage. However, difficulty of upgrade design resides within prediction of future trend such as technological trends and market trends, and upgrade design should be robusts against uncertainty of future predection. This paper proposes a design methodology for upgradable products, and illustrates a prototype upgrade design support tool which can handle changes of user requirement and market trend. With a design example of a vacuume cleaner, this papar also demonstrates the features of the upgrade design tool., The Japan Society of Mechanical Engineers, Japanese - Hypothetical Knowledge Generation Method for Creative Design
YOSHIOKA Masaharu, NOMAGUCHI Yutaka, FUJIMOTO Yutaka, SHIMOMURA Yoshiki, TAKEDA Hideaki, The Proceedings of Design & Systems Conference, 2003, 0, 189, 191, 2003
Since previous designer may already generate possible combination of existing domain knowledge, it is difficult to design creative solution without creating innovative knowledge. We have already confirmed that combination of existing knowledge from different domain may useful for this process by analyzing design cases. In this paper, we propose methodology to generate hypothetical knowledge for creative design by combining existing knowledge., The Japan Society of Mechanical Engineers, Japanese - 設計における創造性支援環境「ユニバーサル・アブダクション・スタジオ」の提案
武田 英明, 坂井 宏充, 野間口 大, 吉岡 真治, 下村 芳樹, 冨山 哲男, 人工知能学会全国大会論文集, 3, 0, 183, 183, 2003
本研究の目的は、設計の創造過程を表現し、計算機上で支援する環境を構築することである。設計の創造的活動はアブダクション、特に理論を統合するアブダクションとしてモデル化することができる。理論を統合するアブダクションの実現には多様なアブダクションのスキーマが併せて利用されるが、ここではアナロジーに基づくアブダクションに注目して、アナロジーに基づく知識統合による設計支援システムのプロトタイプを構築した。, 一般社団法人 人工知能学会, Japanese - 物語性を考慮した要約に基づく類似文書の汎化
原口 誠, 吉岡 真治, 溝江 彰人, 四ッ谷 雅輝, 人工知能学会全国大会論文集, 3, 0, 189, 189, 2003
類似文書群の中で初めて言葉の重要度と関連性が定まるとの立場から、物語性を考慮した文の重要度に基づく文書汎化手法を提案する。また、汎化の品質向上に寄与する辞書の更新についても実験的検当結果を報告する。, 一般社団法人 人工知能学会, Japanese - 問題解決のための知識マネージメントシステムの研究(第2報) - タスクの種類に応じた知識の記述 -
吉岡 真治, 五十嵐 優, 社本 康広, 人工知能学会全国大会論文集, 3, 0, 93, 93, 2003
本研究では、ユーザが持つ問題に対し、類似事例などを提示しながら、問題解決を支援する知識マネージメントシステムの研究を行う。本報では、問題解決のタスクの種類に関する分析と、その分析に応じて適切な知識の構造化手法について提案する。, 一般社団法人 人工知能学会, Japanese - Upgradable Design for Inverse Manufacturing (6th Report)
Yagi Hiroaki, Kondoh Shinsuke, Umeda Yasushi, Shimomura Yoshiki, Yoshioka Masaharu, Proceedings of JSPE Semestrial Meeting, 2003, 0, 257, 257, 2003
製品のアップグレード設計とは、モジュールを追加、交換することにより製品の機能をアップグレードし機能寿命の延命化を図るものであり、機能の陳腐化による廃棄行動を抑制可能である。本稿では、アップグレード時に機能の追加/削除を可能にするアップグレード製品設計の方法論を提案する。具体的には、対象製品の機能·挙動·構造の関係を操作することにより、機能の追加/削除が可能である、追加モジュールおよび世代間で共通のプラットフォームを決定する。, The Japan Society for Precision Engineering - Upgradable Design for Inverse Manufacturing
松田 暁, 下村 芳樹, 吉岡 真治, 梅田 靖, Proceedings of JSPE Semestrial Meeting, 2003, 0, 258, 258, 2003
著者らは,製品の機能的な寿命の延長を実現するための方法であるアップグレード設計方法論の研究を行っている。本研究は、予測された市場の技術動向に対し追従可能であるよう,製品に施す機能的なアップグレードを計画し、その計画を最適に実現可能であるように製品を設計するための方法を与えることを目的とするものであり、製品のアップグレードを効果的に行うためには、製品アップグレードの計画段階でそれによる効果を把握できることが好ましい。本報では、市場に投入される技術の内容、タイミング、製品使用者の指向等に影響される不確実性を考慮した上で、消費者要求の将来予測を行い、それに基づいた製品ラインナップに要求される仕様を把握する方法について提案する。, The Japan Society for Precision Engineering - Development of Universal Abduction Studio (1st Report)
下村 芳樹, 坂井 宏充, 野間口 大, 吉岡 真治, 武田 英明, 冨山 哲男, Proceedings of JSPE Semestrial Meeting, 2003, 0, 27, 27, 2003
著者らは設計における創造性の理解とその計算機による支援を目的とする研究を行っており、その一環として統合的推論環境Universal Abduction Studioの開発に着手している。創造性の高い設計支援を実現する一つの鍵は、対象固有の領域知識のみでなく、異なる領域に属する知識を横断的に利用可能とすることであると考えられる。本研究では、異領域の知識を横断的に利用する推論形態としてのAbductionに着目し、これを統合的に実装することにより創造性の高い設計支援のための推論環境を提供するUniversal Abduction Studioの提案を行う。, The Japan Society for Precision Engineering - Development of Universal Abduction Studio (2nd Report)
Sakai Hiromitsu, Nomaguchi Yutaka, Shimomura Yoshiki, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2003, 0, 28, 28, 2003
著者らは設計における創造性の理解とその計算機による支援を目的に研究を行っている。創造性の高い設計の支援を考える場合、対象固有の領域知識に留まらず、異なる領域に属する知識を横断的に利用可能とすることが必要である。本研究では、知識間の構造的な類似性を発見する枠組みと、これを創造性の高い設計支援に適用する手法を提案する。特に、知識間の類似性発見の手法について、例題よりその有効性を検証する。, The Japan Society for Precision Engineering - Research on Knowledge Management System for Problem Solving (1st Report) Knowledge Description based on Task Type
吉岡 真治, 五十嵐 優, 社本 康広, 人工知能学会全国大会論文集, 17, 1, 3, 2003
人工知能学会, Japanese - Abstracting Similar Documents based on a Text Summarization Technique
原口 誠, 吉岡 真治, 溝江 彰人, 人工知能学会全国大会論文集, 17, 1, 3, 2003
人工知能学会, Japanese - Development of Universal Abduction Studio (3rd report)
Fujimoto Yutaka, Nomaguchi Yutaka, Shimomura Yoshiki, Yoshioka Masaharu, Takeda Hideaki, Proceedings of JSPE Semestrial Meeting, 2003, 0, 39, 39, 2003
対象固有の領域知識を用いるだけでは、既存の設計解を改良することは可能であっても創造性の高い設計を支援することは困難である。即ち、創造性の高い設計を支援するためには、異なる領域に属する知識を関連づけることによる問題の再定義や、新しい知識を作成するための手段が必要である。本報では、前報で提案した知識間の類似性を発見する枠組とその評価結果に基づき、より高度な知識利用を実現するための知識表現の枠組みに対する要求仕様を明らかにするとともに、その具体的な実現方法に関して考察する。, The Japan Society for Precision Engineering, Japanese - AI in Design Webliography
吉岡 真治, Journal of Japanese Society for Artificial Intelligence, 18, 1, 101, 101, 01 Jan. 2003
社団法人人工知能学会, Japanese - Research Index : 新しい形の電子図書館(Web探訪)
吉岡 真治, 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 17, 5, 649, 650, 01 Sep. 2002
社団法人人工知能学会, Japanese - Organization of Documents in Terms of Maximal Analogies
HARAGUCHI Makoto, NAKANO Shigetora, YOSHIOKA Masaharu, IPSJ SIG Notes, 150, 66, 99, 103, 15 Jul. 2002
Given two or more similar documents in the form of texts, we present a notion of maximal analogies representing maximal sequences consisting of pairs of similar events in the documents. They are required to satisfy certain cost condition so that meaningless similarities between documents are never concluded. A bottom-up search procedure to find a maximal analogy satisfying the cost condition is also presented. In addition to the set of similar documents, we suppose another set of dissimilar ones. Then a maximal analogy is furthermore tested for their appropriateness so as not to explain the latter ones. The test can be performed by an effective subsumption check procedure., Information Processing Society of Japan (IPSJ), Japanese - Organization of Documents in Terms of Maximal Analogies
HARAGUCHI Makoto, NAKANO Shigetora, YOSHIOKA Masaharu, IEICE technical report. Natural language understanding and models of communication, 102, 200, 1, 5, 09 Jul. 2002
Given two or more similar documents in the form of texts, we present a notion of maximal analogies representing maximal sequences consisting of pairs of similar events in the documents. They are required to satisfy certain cost condition so that meaningless similarities between documents are never concluded. A bottom-up search procedure to find a maximal analogy satisfying the cost condition is also presented. In addition to the set of similar documents, we suppose another set of dissimilar ones. Then a maximal analogy is furthermore tested for their appropriateness so as not to explain the latter ones. The test can be performed by an effective subsumption check procedure., The Institute of Electronics, Information and Communication Engineers, Japanese - Research on Information Retrieval System based on Adaptive Generalization (1st Report) : Generalization of Keywords on the Analysis on Contribution of Keyword based for Relevance Judgement
YOSHIOKA Masaharu, HARAGUCHI Makoto, OHKUBO Yoshiaki, IPSJ SIG Notes, 127, 41, 151, 158, 21 May 2002
It is not easy for a user of Information Retrieval (IR) system to select appropriate keywords. Therefore, many IR systems have capability to modify keywords by estimating user's intent. However, since modified keywords are usually represented as complicated form, it is difficult to judge the appropriateness of them. In this research, we proposed a new IR system that uses adaptive generalization of keywords. when the system can selects appropriate generalization by estimating user's intent, the system can generate good keywords that have high readability and good retrieval performance. In this report, we proposed to use the contribution value of each keyword for relevance judgement to select appropriate generalization. In addition, we confirm general concept structure stored in a thesaurus is not sufficient for representing particular user's intent. Therefore, we proposed a purpose oriented method to modify concept structure., Information Processing Society of Japan (IPSJ), Japanese - Research on Information Retrieval System based on Adaptive Generalization (1st Report) - Generalization of Keywords based on the Analysis on Contribution of Keyword for Relevance Judgement-
吉岡 真治, 原口 誠, 大久保 好章, 情報処理学会研究報告情報学基礎(FI), 2002, 41, 151, 158, 21 May 2002
情報検索システムを利用する検索者にとって、適切なキーワードを選択することは必ずしも容易なことではない。この問題に対し、検索者の検索意図を推定し、検索者の補完を行う検索システムが提案されている。しかし、補完した結果は複雑であることが多く、検索者が補完結果自体を評価することが困難である。本研究では、検索者にも理解しやすい概念階層の汎化という考え方を利用して、ユーザの検索意図を明示化すると共に、精度落ちを抑えた情報検索システムを提案する。本報では、概念階層の汎化のレベルを決定するために、検索者により入力された検索語が持つ適合性判定への寄与度を考える。また、本システムの性能を向上させるためには、概念階層自体が目的に応じて再構成されることが必要であることについて述べ、目的指向の概念階層の修正の方法を提案する。It is not easy for a user of Information Retrieval (IR) system to select appropriate keywords. Therefore, many IR systems have capability to modify keywords by estimating user's intent. However, since modified keywords are usually represented as complicated form, it is difficult to judge the appropriateness of them. In this research, we proposed a new IR system that uses adaptive generalization of keywords. When the system can selects appropriate generalization by estimating user's intent, the system can generate good keywords that have high readability and good retrieval performance. In this report, we proposed to use the contribution value of each keyword for relevance judgement to select appropriate generalization. In addition, we confirm general concept structure stored in a thesaurus is not sufficient for representing particular user's intent. Therefore, we proposed a purpose oriented method to modify concept structure., Japanese - Analysis of Japanese Deverbal Compounds Based on Lexical Conceptual Structure(
Design Technologies and Design Automation of Electronic Systems)
TAKEUGHI KOICHI, UCHIYAMA KIYOKO, YOSHIOKA MASAHARU, KAGEURA KYO, KOYAMA TERUO, IPSJ Journal, 43, 5, 1446, 1456, 15 May 2002
In this paper, we describe a principled approach for analyzing relations between constituent words of compound nouns, specifically those whose heads are deverbal nouns, based on the classification of deverbal nouns by their lexical conceptual structure (LCS) and the classification of nouns in modifier position vis-a-vis LGS of head deverbal nouns. There are two kinds of relations of compounds with head deverbal nouns. The one is that a modifier noun becomes an argument of deverbal head and the other is that a modifier becomes an adjunct. It is an important starting point for analyzing relat..., Information Processing Society of Japan (IPSJ), Japanese - Editor's Introduction to "NTCIR from the View Point of Participant"(
NTCIR from the View Point of Participant)
Masaharu Yoshioka, Noriko Kando, Journal of Japanese Society for Artificial Intelligence, 17, 3, 295, 295, 01 May 2002
社団法人人工知能学会, Japanese - AI Topics (AAAI)
吉岡 真治, 人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 17, 3, 372, 373, 01 May 2002
社団法人人工知能学会, Japanese - Construction of a Large Scale Test Collection NTCIR-2 : The Effect of Additional Interactive Search and Cross-Lingual Pooling(<特集>Information Retrieval Technology and Test Collections)
KURIYAMA KAZUKO, YOSHIOKA MASAHARU, KANDO NORIKO, 情報処理学会論文誌. データベース, 43, 2, 48, 59, 15 Mar. 2002
The purposes of this study are to examine whether there is an effect on the relative evaluation of the IR systems using the relevance judgments of the test collection NTCIR-2 made by the pooling method and additional interactive searches, and to investigate whether the NTCIR-2 is effective for evaluating the IR systems, the search results of which were not used for the pooling. We carried out experiments using different lists of relevance judgments and search results submitted for the test of the 2nd NTCIR Workshop. First, we evaluated the search results using the list of the final relevanc..., Information Processing Society of Japan (IPSJ), Japanese - インパース・マニュファクチャリングのためのアップグレーダブル設計(第5報):不確実性に対応した設計支援ツールの開発
梅森 嘉匡, 八木 宏明, 近藤 伸亮, 梅田 靖, 下村 芳樹, 吉岡 真治, 精密工学会大会学術講演会講演論文集, 2002, 1, 138, 138, 01 Mar. 2002
Japanese - 文書情報と形式化された知識を融合する知識マネージメントの研究 (第1報)-文書情報と形式的知識の融合-
吉岡真治, 2002年度人工知能学会全国大会講演論文集, 2002 - AI Topics (AAAI)
吉岡 真治, Journal of the Japanese Society for Artificial Intelligence, 17, 3, 372, 373, 2002
The Japanese Society for Artificial Intelligence, Japanese - Editor's Introduction to "NTCIR from the View Point of Participant"(
NTCIR from the View Point of Participant)
YOSHIOKA Masaharu, KANDO Noriko, Journal of the Japanese Society for Artificial Intelligence, 17, 3, 295, 295, 2002
The Japanese Society for Artificial Intelligence, Japanese - NTCIR from Participants' Viewpoints(
NTCIR from the View Point of Participant)
OGAWA Yasushi, SASAKI Yutaka, MASUYAMA Shigeru, MURATA Masaki, YOSHIOKA Masaharu, Journal of the Japanese Society for Artificial Intelligence, 17, 3, 306, 311, 2002
The Japanese Society for Artificial Intelligence, Japanese - Development of Upgrage Design Support System for Inverse Manufacturing
ISHIGAMI Yoshiaki, KONDOH Shinsuke, UMEDA Yasushi, SHIMOMURA Yoshiki, YOSHIOKA Masaharu, The Proceedings of Design & Systems Conference, 2002, 0, 56, 59, 2002
Upgrade design is one of the usefull approaches to realize inverse manufacturing because upgrade design can extend product lives and reduce consumption volume of resources and energy usage. However, difficulty of upgrade design resides within prediction of future trend such as technological trends and market trends. This papar propose a design method for upgradable products, and prototype design support tool which can be adapted changes of user requirement and market trend. Using design example of vacuume cleaner, this papar demonstrates that this support tools is effective for designing upgradable products., The Japan Society of Mechanical Engineers, Japanese - Advances in Research into Artifacts (18^
Report) : Proposal of the Universal Abduction Studio
SAKAI Hiromitsu, NOMAGUCHI Yutaka, YOSHIOKA Masaharu, TAKEDA Hideaki, SHIMOMURA Yoshiki, TOMIYAMA Tetsuo, The Proceedings of Design & Systems Conference, 2002, 0, 410, 411, 2002
Although design support systems have been well developed in geometric and detail design areas, they are still not successful in conceptual or creative design domain. In our opinion, the main difficulty is knowledge and its operation that play a crucial role in conceptual design. In this paper, we propose "Universal Abduction Studio" that is a new approach to support designers by computers, and discuss technologies to realize it., The Japan Society of Mechanical Engineers, JapaneseResearch on Knowledge Management System for Problem Solving (1st Report):-Integration of Document information and Formalized Knowledge-
Yoshioka Masaharu, Igarashi You, Shamoto Yasuhiro, Proceedings of the Annual Conference of JSAI, 2, 0, 142, 142, 2002
The Japanese Society for Artificial IntelligenceResearch on Knowledge Management System for Problem Solving(1st Report)Integration of Document information and Formalized Knowledge
吉岡 真治, 五十嵐 優, 社本 康広, 人工知能学会全国大会論文集, 16, 1, 4, 2002
人工知能学会, JapaneseA Proposal of CAD Mechanism for Design Knowledge Management
Nomaguchi Yutaka, Yoshioka Masaharu, Tomiyama Tetsuo, Transactions of the Japanese Society for Artificial Intelligence, 17, 17, 104, 113, 2002
In this paper, we propose a fundamental idea of a new CAD mechanism to facilitate design knowledge management. This mechanism encourages a designer to externalise his/her knowledge during a design process and facilitates sharing and reuse of such externalised design knowledge in later stages. We also describe the implementation of this idea called DDMS (Design Documentation Management System). DDMS works as a front end to KIEF (Knowledge Intensive Engineering Framework), which we have been developing. We also illustrate an example of machining tool design to demonstrate the features of DDMS., The Japanese Society for Artificial Intelligence, JapaneseConstruction of a Large Scale Test Collection NTCIR-2 : The Effect of Cross-Lingual Pooling on Evaluation
KURIYAMA Kazuko, YOSHIOKA Masaharu, KANDO Noriko, IPSJ SIG Notes, 63, 74, 9, 16, 24 Jul. 2001
The purpose of this study is to examine whether there is an effect on the relative evaluation of the IR systems using the relevance judgments made by the pooling method and additional interactive searches. We carried out experiments using different lists of relevance judgments and search results submitted for the test of the 2nd NTCIR Workshop. First, we evaluated the search results using the list of the final relevance judgments F of NTCIR-2 and F-I, that is, the F without the unique relevant documents found by the additional interactive searches I. Second, we made pools from the search results for each of the sub-tasks and evaluated the search results using the pools as lists of relevance judgments. Almost the same rankings of the search results were produced by using the pools as lists of relevance judgments for system evaluation. Therefore our results verified the reliability of test collection as an evaluation tool, which was based on pooling method., Information Processing Society of Japan (IPSJ), JapaneseInformation Retrieval System Evaluation using Multi-grade Relevance Judgments : Discussion on Averageable Single-numbered Measures
KANDO Noriko, KURIYAMA Kazuko, YOSHIOKA Masaharu, IPSJ SIG Notes, 63, 74, 105, 112, 24 Jul. 2001
This paper proposes new measures, weighted R precision and wighted average precision, for information retrieval(IR) system testing with ranked output based on multi-grade relevance judgments. Relevance judgments are one of the most fundamental factors in evaluation of search effectiveness of IR systems. Intuitively, the highly relevant documents are more important for users than partial relevant ones and the documents retrieved in the higher ranks in the ranked list are more important. Therefore the systems producing the search results in which higher relevant documents in higher ranks in the ranked list should be rated as better thought relevance judgments must be done in binary to calculate traditional IR system evaluation measures "recall" and "precision". Based on the review of existing IR system evaluation measures, we decided that either of proposed measures is single number and averageable over number of runs. The paper then shows the evaluation results of the Japanese IR task at the second NTCIR Workshop, in which four-grade relevance judgments, highly relevant, relevant, partially relevant, and irrelevant, were used. Finally implications of "partial relevance" are discussed in the relation to user-oriented IR system evaluation studies., Information Processing Society of Japan (IPSJ), JapaneseThe effect of cross-lingual pooling on evaluation
Kuriyama,K, Yoshioka, M, Kando, N, Proceedings of the Second NTCIR Workshop on Research in Chinese & Japanese Text Retrieval and Text Summarization., Tokyo, Japan, May 2000-March 2001, National Institute of Informatics, 297, 310, Jul. 2001
English, Book reviewOverview of Japanese and English Information Retrieval Tasks (JEIR) at the Second NTCIR Workshop
Kando, N, Kuriyama, K, Yoshioka, M, Proceedings of the Second NTCIR Workshop on Research in Chinese & Japanese Text Retrieval and Text Summarization, Tokyo, Japan, May 2000-March 2001, National Institute of Informatics, 73, 96, Jul. 2001
English, Book reviewAnalysis of the usage of Japanese segmented texts in NTCIR Workshop 2
Yoshioka, M, Kuriyama, K, Kando, N, Proceedings of the Second NTCIR Workshop on Research in Chinese & Japanese Text Retrieval and Text Summarization., Tokyo, Japan, May 2000-March 2001, National Institute of Informatics, 291, 296, Jul. 2001
English, Book reviewインバース・マニュファクチャリングのためのアップグレーダブル設計 (第4報) アップグレーダブル設計におけるプラニング戦略
下村芳樹, 吉岡真治, さい剛民, 梅森嘉匡, 梅田靖, 精密工学会大会学術講演会講演論文集, 2001, 365, 05 Mar. 2001
Japanese人と情報システムとのシンビオシスのための情報検索および情報活用支援に関する研究 -- Web文書検索を中心として--
栗山和子, 吉岡真治, 江口浩二, 岸田和明, 神門典子, 影浦峡, 高野明彦, 大山敬三, 安達淳, COEプロジェクト平成12年度報告書『シンビオティック情報システムへの概念形成と実現技術』, 国立情報学研究所,2001, 47, 54, Mar. 2001
Japanese, Book reviewA proposal of design reasoning model that takes notice of design knowledge (the Third Report) : A proposal of a synthesis language for describing design operational knowledge
YOSHIOKA Masaharu, Takeda Hideaki, WASHIO Takashi, TOMIYAMA Tetsuo, The Proceedings of Design & Systems Conference, 2001, 0, 281, 284, 2001
Since design knowledge plays a crucial role in design process, we are conducting research on design reasoning model that takes notice of design knowledge. To make a design reasoning model, we have already proposed a reasoning framework of design and analyzed basic operations for the framework. However, since these operations are just fragments of whole design process, it is necessary to describe design operational knowledge that can control the entire reasoning operations by constructing the sequence of these fragments. So, in this paper, we propose a synthesis language for describing design operational knowledge., The Japan Society of Mechanical Engineers, JapaneseA proposal of design reasoning mode that takes notice of design knowledge (the Forth report) : Implementation of a reasoning model in design and its verification
Nomaguchi Yutaka, Tsumaya Akira, Yoshioka Masaharu, Washio Takashi, Takeda Hideaki, Murakami Tamotsu, Tomiyama Tetsuo, The Proceedings of Design & Systems Conference, 2001, 0, 285, 288, 2001
Since design knowledge plays a crucial role in design process, we are conducting research on design reasoning model that takes notice of design knowledge. To make a design reasoning model, we have already proposed a reasoning framework of design and analyzed basic operations for the framework, and then propose a synthesis language for describing design operational knowledge. In this paper, we propose our implementation of this model, and verify it to compare design processes replayed in the system with ones of actual design sessions., The Japan Society of Mechanical Engineers, JapaneseA Proposal of Design Reasoning Model that Takes Notice of Design Knowledge (the Fifth Report) : Comparison of a reasoning model with Existing Design Process Models
TSUMAYA Akira, NOMAGUCHI Yutaka, YOSHIOKA Masaharu, TAKEDA Hideaki, MURAKAMI Tamotsu, TOMIYAMA Tetsuo, The Proceedings of Design & Systems Conference, 2001, 0, 289, 292, 2001
In engineering design research, numerous design process models have already been proposed. For constructing model of synthesis, it is very important to know the position of the model among them. This paper reports results of comparison of some design process models with the synthesis oriented thought process model. We use a reference model as an example of design, compare it with some existing design process models, and the results of this analysis are described., The Japan Society of Mechanical Engineers, Japanese206 A Proposal of CAD Mechanism for Design Knowledge Management
NOMAGUCHI Yutaka, YOSHIOKA Masaharu, TOMIYAMA Tetsuo, 設計シンポジウム講演論文集, 2001, 19, 109, 116, 2001
In this paper, we propose a fundamental idea of a new CAD architecture to facilitate design knowledge management. This architecture is based on knowledge-based design environment for encouraging the designer to externalize his/her knowledge. We also discuss the implementation of this idea. DDMS (Design Documentation Management System) works as a front end for KIEF (Knowledge Intensive Engineering Framework), which we have been developing. We also illustrate an example of machining tool design to demonstrate the system's features., 社団法人日本機械学会, Japanese2314 Advances in Research into Artifacts (The 15th Report) : Knowledge Acquisition by Discovering the Designer's Thought Patterns
Sakai Hiromitsu, Nomaguchi Yutaka, Yoshioka Masaharu, Shimomura Yoshiki, Tomiyama Tetsuo, The Proceedings of Design & Systems Conference, 2001, 0, 204, 206, 2001
In engineering activities, management of designers' knowledge is becoming crucial to share and reuse it efficiently. Toward this issue, we are studying a methodology to capture the designer's thought process, and to semi-automatically arrange it into a refined document, which addresses a particular objective with particular thought patterns. The key issues of this methodology are to acquire the thought patterns and to define relations among them. This paper reports our algorithm to acquire thought patterns from a record of designer's thought processes and any other types of documents including patent documents and academic papers. We employ a knowledge discovery approach which is recently discussed in data mining research., The Japan Society of Mechanical Engineers, Japanese3104 Design for Upgradable Products Considering Future Uncertainty
UMEMORI Yoshimasa, KONDOH Shinsuke, UMEDA Yasushi, SHIMOMURA Yoshiki, YOSHIOKA Masaharu, The Proceedings of Design & Systems Conference, 2001, 0, 269, 272, 2001
Conventional products tend to be thrown away because of functional obsoleteness before reaching the end of physical life of the products. This is one of the main causes of the mass disposal problem. In order to solve this problem, this paper proposes a design methodology for upgradable products. Especially, this paper discusses a method dealing with uncertainty of long-term planning in the design methodology. The proposed method deals with uncertainty as ranges of parameter values, and derives a design solution that realizes required upgrades, while adapting to estimated uncertainty. This method makes a design object robust and tolerant against the uncertainty and, therefore, makes design for upgradability more feasible., The Japan Society of Mechanical Engineers, Japanese3305 Research on knowledge management system that harmonize the document information and well formalized knowledge
YOSHIOKA Masaharu, SHAMOTO Yasuhiro, The Proceedings of Design & Systems Conference, 2001, 0, 321, 324, 2001
In engineering activities, engineers solve various kinds of problems. Most of these problems are routine problems that are easy to formalize and formalized knowledge is useful for these problems. On the other hand, there are complicated problems that is difficult to formalize all of the problem solving process. One of the useful information to solve such complicated problems is result of the similar problem solving process. In addition, later in such complicated problem solving process, most of the decomposed subproblems are able to handle as routine problems. In this research, we decide to use formalized knowledge and document information that represents the history of problem solving processes to support to solve both types of the problem. For that purpose, we propose a new knowledge management system that supports a user to solve the problem through document processing. In this support process, the system retrieves related documents for user's problems by using formalized knowledge and information retrieval techniques. After retrieving related documents, he/she modifies them to fit his/her problems. Modified information is associated with the formalized knowledge and stored as documents for future reuse., The Japan Society of Mechanical Engineers, JapaneseRoles and formalization of abduction in synthesis
Takeda Hideaki, Yoshioka Masaharu, Tomiyama Tetsuo, Proceedings of the Annual Conference of JSAI, 1, 0, 30, 30, 2001
The Japanese Society for Artificial IntelligenceRoles and formalization of abduction in synthesis
武田 英明, 吉岡 真治, 冨山 哲男, 人工知能学会全国大会論文集, 15, 1, 4, 2001
人工知能学会, JapaneseLarge Scale Test Collections NTCIR-1 and NTCIR2 : Differences in the Document Records
KURIYAMA Kazuko, YOSHIOKA Masaharu, KANDO Noriko, IPSJ SIG Notes, 60, 91, 9, 16, 27 Sep. 2000
In this paper we describe large scale test collections, NTCIR-1 and NTCIR-2, and the 2nd NTCIR Workshop. We also show differences in the records of the test collections from point of view of statistical data ; length of the records, their distribution, number of different terms, their frequencies, and so on., Information Processing Society of Japan (IPSJ), JapaneseLarge Scale Test Collections NTCIR-1 and NTCIR2 - Differences in the Document Records -
Kuriyama Kazuko, Yoshioka Masaharu, Kando Noriko, Digital libraries, 18, 18, 57, 64, Sep. 2000
図書館情報大学, Japanese〈発表論文〉大規模テキストコレクションNTCIR-1とNTCIR-2 : レコードから見た違い(情報処理学会 情報学基礎研究会(FI)第60回)
栗山和子, 吉岡真治, 神門典子, Kuriyama kazuko, Yoshioka Masaharu, Kando Noriko, クリヤマカズコ, ヨシオカマサハル, カンドノリコ, ディジタル図書館, 18, 57, 64, Sep. 2000
ディジタル図書館編集委員会設計活動を表現する語彙に関する研究(第3報) : 情報操作の視点から見た語彙の整理と意味付け
吉岡 真治, 武田 英明, 妻屋 彰, 村上 存, 冨山 哲男, 精密工学会大会学術講演会講演論文集, 2000, 2, 259, 259, 01 Sep. 2000
Japanese設計事例分析による設計過程モデルの比較(第2報) : 設計過程の情報モデルの提案
野間口 大, 妻屋 彰, 吉岡 真治, 武田 英明, 村上 存, 冨山 哲男, 精密工学会大会学術講演会講演論文集, 2000, 2, 260, 260, 01 Sep. 2000
JapaneseResearch on Reasoning Framework for Synthesis (The First Report) : The Basic Requirements for Modeling of Synthesis
TAKEDA Hideaki, YOSHIOKA Masaharu, TOMIYAMA Tetsuo, Proceedings of the Annual Conference of JSAI, 14, 0, 464, 467, 03 Jul. 2000
人工知能学会, JapaneseReseach on Reasoning Framework for Synthesis (The Second Report) : Model-based Abduction
YOSHIOKA Masaharu, TAKEDA Hideaki, TOMIYAMA Tetsuo, Proceedings of the Annual Conference of JSAI, 14, 0, 468, 471, 03 Jul. 2000
人工知能学会, JapaneseBuilding Japanese Compound Analyzer Using Lexical Restriction
TAKEUCHI Koichi, UCHIYAMA Kiyoko, YOSHIOKA Masaharu, KAGEURA Kyo, KOYAMA Teruo, IPSJ SIG Notes, 57, 29, 71, 78, 21 Mar. 2000
The aim of our research is to build a Japanese compound analyzer using grammatical constraint. To analyze the compounds which have verbs in their heads is to reveal the relationship between incorporated nouns and the head verbal nouns. The kind of relationship is adjunctive and internal argument of a head verb. In our previous research, we categorize verbs based on their argument structure and lexical conceptual structure (LCS), but it is necessary to categorize nouns to reveal the relationship between a noun and a head verb in compounds. Therefore we propose the method to categorize nouns based on argument structure and LCS of head verbs in this paper. In the case of categorization, we also use qualia structure of generative lexicon. After the assumptions of restriction about categorized nouns is constructed, we attempt the experiment of analyzing technical terms (103 words) in the domain of information processing. Finally we show the experimental results that 92 words (89%) can be solved using our restriction and discuss the limitation of our model., Information Processing Society of Japan (IPSJ), JapaneseBuilding Japanese Compound Analyzer Using Lexical Restriction
Takeuchi Koichi, Uchiyama Kiyoko, Yoshioka Masaharu, Kageura Kyo, Koyama Teruo, IPSJ SIG Notes, 2000, 29, 71, 78, 21 Mar. 2000
The aim of our research is to build a Japanese compound analyzer using grammatical constraint. To analyze the compounds which have verbs in their heads is to reveal the relationship between incorporated nouns and the head verbal nouns. The kind of relationship is adjunctive and internal argument of a head verb. In our previous research, we categorize verbs based on their argument structure and lexical conceptual structure (LCS), but it is necessary to categorize nouns to reveal the relationship between a noun and a head verb in compounds. Therefore we propose the method to categorize nouns based on argument structure and LCS of head verbs in this paper. In the case of categorization, we also use qualia structure of generative lexicon. After the assumptions of restriction about categorized nouns is constructed, we attempt the experiment of analyzing technical terms (103 words) in the domain of information processing. Finally we show the experimental results that 92 words (89%) can be solved using our restriction and discuss the limitation of our model., Information Processing Society of Japan (IPSJ), JapaneseBuilding Japanese Compound Words Analyzer Based on Grammatical Constraint
Takeuchi Koichi, Uchiyama Kiyoko, Yoshioka Masaharu, Kageura Kyo, Koyama Teruo, Research bulletin of the National Center for Science Information Systems, 12, 12, 7, 15, Mar. 2000
In the previous research of analyzing compound words, the technique using statistics of word co-occurrence is well known. However, since some grammatical and semantic restrictions must control the construction of compound words, only the statistical technique is not enough to analyze compound words. In this paper, we analyze technical terms that consist of two nouns whose head is verbal noun by means of the rich lexical knowledge that are a level ordering assumption, an argument structure and a lexical conceptual structure used on morphology or syntax. We will show the following three results: 1) there are only two kinds of relationship that is a modifier or a argument for a verbal noun in the relationship between two words in a compound word, 2) the analysis of compound words depends on argument structure of verbal noun, and 3) the structure of some compound words are changed according to the context. Finally we discuss constructing compound words analyzer from the results., National Institute of Informatics, Japanese設計活動を表現する語彙に関する研究(第1報) : 語彙の階層的整理
吉岡 真治, 妻屋 彰, 冨山 哲男, 武田 英明, 精密工学会大会学術講演会講演論文集, 2000, 1, 18, 18, 01 Mar. 2000
精密工学会, Japanese設計活動を表現する語彙に関する研究(第2報) : シンセシスのフレームワークとの対応
武田 英明, 吉岡 真治, 妻屋 彰, 冨山 哲男, 精密工学会大会学術講演会講演論文集, 2000, 1, 19, 19, 01 Mar. 2000
精密工学会, Japanese設計事例分析による設計過程モデルの比較(第1報) : 設計事例記録の整理
妻屋 彰, 野間口 大, 吉岡 真治, 武田 英明, 村上 存, 冨山 哲男, 精密工学会大会学術講演会講演論文集, 2000, 1, 20, 20, 01 Mar. 2000
Japaneseインバース・マニュファクチャリングのためのアップグレーダブル設計(第3報) : 支援システムの試作
今川 晴紀, 下村 芳樹, 吉岡 真治, 梅田 靖, 精密工学会大会学術講演会講演論文集, 2000, 1, 96, 96, 01 Mar. 2000
JapaneseDesign for Upgradable Aimed to the Inverse Manufacturing. (2nd Report). Identification of Upgrading Functions Using Design Object Models and Function Modifiers.
下村芳樹, 今川晴紀, 吉岡真治, 梅田靖, 日本機械学会設計工学・システム部門講演会講演論文集, 9th, 501, 504, 19 Nov. 1999
JapaneseAnalysis of Japanese Compound Words Based on Rich Lexical Knowledge
TAKEUCHI Koichi, UCHIYAMA Kiyoko, YOSHIOKA Masaharu, KAGEURA Kyo, KOYAMA Teruo, IEICE technical report. Natural language understanding and models of communication, 99, 387, 7, 14, 25 Oct. 1999
It is difficult problem to know how to prepare lexical knowledge in order to analyze Japanese compound words. In this paper, we analyze technical terms that consist of two verbal nouns by means of the rich lexical knowledge that are a level ordering assumption, an argument structure and a lexical conceptual structure used on morphology or syntax. Our analysis of compound words aims to reveal the relations between the constituent words of compound words. From the results, our method can explain some parts of phenomena of the impossible compound words and the conversion of verbal noun into noun in compound words. The lexical knowledge we have assumed is aimed at describing the details of verbal nouns. While analyzing verbal nouns, we also clarify how to describe the lexical knowledge of noun words., The Institute of Electronics, Information and Communication Engineers, JapaneseReserach on Design Documentation Management System
Yoshioka Masaharu, Nomaguchi Yutaka, Tomiyama Tetsuo, IPSJ SIG Notes, 99, 88, 63, 68, 21 Oct. 1999
In a design process, the designer uses a variety of information and knowledge for generating design solution and evaluating it. These types of process information are useful not only for understanding the design rationale but also for the source of designers' knowledge. Therefore, many companies try to store these types of information as design documents. However, since most of the documents are made for explaining the design solution after finishing whole design, most of the documents do not cover all of information in design process. In addition, there is not good framework to retrieve and reuse these documents and that prevents the designers to reuse these useful documents. For tackling these problems, we have proposed the design documentation management system. This system can record design documents as a result of design process. The system can also retrieve and reuse related design documents based on the design context. In this report, we discuss the Japanization of the design documentation system and application of text retrieval technique., Information Processing Society of Japan (IPSJ), JapaneseA Study for Connections among Constituent Elements of Nominal Compounds : An Evaluation of Grammatical Categories based on Grammatical Characteristics
UCHIYAMA Kiyoko, TAKEUCHI Koichi, YOSHIOKA Masaharu, KAGEURA Kyo, KOYAMA Teruo, Technical report of IEICE. Thought and language, 99, 237, 7, 12, 26 Jul. 1999
Technical compound nouns are often used as keywords for information retrieval. The mechanism how nominal compounds are generated can be clarified by analyzing internal structures of nominal compounds. The conventional approach has centered on the analysis of semantic relations among constituent elements of nominal compounds. We think structural and grammatical characteristics play the key role in nominal compounds. In our study, we propose original categories in terms of structural and grammatical characteristics, and examine modifications and connections among constituent elements of compound words, using the original categories., The Institute of Electronics, Information and Communication Engineers, JapaneseResearch on Keyword Role Analysis based on the information of title and abstract
Yoshioka Masaharu, Research bulletin of the National Center for Science Information Systems, 11, 11, 27, 31, Mar. 1999
There are several types of research articles; e.g., survey articles, those that propose new technologies, those that apply existing technologies to unsolved problems. In this paper, I propose a methodology to categorize these types of research articles by analyzing roles of the author added keywords. First, I categorize verbs that are frequently used in research articles into five role categories. Second, I propose methodology to analyze roles of the keywords by using the categories of verbs. To validate this methodology, I apply it to the real research articles and explain its result., National Institute of Informatics, JapaneseGrammatical Categories for Constituent Elements of Japanese Nominal Compounds with Special Reference to Technical Terms
Uchiyama Kiyoko, Takeuchi Koichi, Yoshioka Masaharu, Kageura Kyo, Koyama Teruo, Research bulletin of the National Center for Science Information Systems, 11, 11, 49, 57, Mar. 1999
Grammatical information for constituent elements of nominal compounds are needed for analyzing nominal compounds, especially those of technical fields. In this research we analyzed the grammatical categories suitable for nominal compound analysis, as it is difficult to apply conventional grammatical categories defined with respect to words in sentences to the analysis of nominal compounds., National Institute of Informatics, JapaneseConstruction of Terminologically-motivated Corpus
YOSHIOKA Masaharu, OKADA Maho, KAGEURA Kyo, KOYAMA Teruo, IPSJ SIG Notes, 53, 20, 41, 48, 01 Mar. 1999
We are currently constructing the terminology-motivated corpus (the NACSIS corpus) as a part of the NACSIS test collection project. There are two issues to constructing this corpus. One is a definition of domain that makes the terminology research different from IR keyword one. The other is a definition of lexical units in the texts that gives basic unit for terminology operations. In this paper, we discuss the approach to design the NACSIS corpus for solving these two issues, and show the status of NACSIS., Information Processing Society of Japan (IPSJ), JapaneseThe Development of Knowledge Integrated Engineering Framework Based on Ontology ("Ontology : Foundations and Applications")
SEKIYA Takayuki, YOSHIOKA Masaharu, TOMIYAMA Tetsuo, Journal of the Japanese Society for Artificial Intelligence, 14, 6, 1051, 1060, 1999During design, a designer uses various computational tools, such as a geometric modeling system, analysis tools, and databases. To support these design processes, a system that can integrate such computational tools is required. For this purpose, we have been developing the Knowledge Intensive Engineering Framework (KIEF) system. KIEF enables the designer to build, evaluate, and modify design models on the system by providing him/her with a wide variety of engineering knowledge. A common ontology plays an important role in integrating tools and representing engineering knowledge. This paper describes the structure of the ontology in KIEF, and how it is used in a modeling processes on KIEF.
, The Japanese Society for Artificial Intelligence, JapaneseAplication of Knowledge Intensive Engineering to Architectural Design
Yoshioka Masaharu, Research bulletin of the National Center for Science Information System, 10, 0, 73, 79, Mar. 1998
Because of the finiteness of natural resources, environmental problems and so on, engineers are expected to evaluate their products from ever more various kinds of aspects. To support such claim, we proposed the concept of knowledge intensive engineering, in which various kinds of knowledge is used in a flexible and integrated manner in order to aim at generating more added-value. In addition, we also proposed Knowledge Intensive Engineering Framework (KIEF) system that forms a computational framework of knowledge intensive engineering. In this paper, I describe the concept of knowledge intensive engineering and the KIEF system. After that, I apply the KIEF system to architectural design to show the value of the system. To do so, I analyze knowledge that is used in architecture design and discuss how to implement the knowledge on the KIEF system. Finally, I show some results of this application and discuss the capability of the KIEF system., National Institute of Informatics, JapaneseA Study on an Integrated Modeling Environment for Design: A Proposal of the Pluggable Metamodel Mechanism
YOSHIOKA Masaharu, TOMIYAMA Tetsuo, Journal of the Japanese Society for Artificial Intelligence, 13, 2, 312, 319, 1998In this study, we propose a framework of an integrated modeling environment for design. Since a mechanical engineering design process requires various kinds of design object models, such as a geometric model, a kinematic model, and a finite element model, this environment should maintain consistency among these models. Furthermore, commercial design tools are practically used in design processes, and these must be integrated as well. For dealing with multiple design object models, we have formalized a concept called a metamodel. A metamodel is a model which represents relationships among concepts used in various design object models. This paper proposes a framework of the pluggable metamodel mechanism that allows to plug in existing design tools, to support a modeling on it, and to maintain consistency among them. We describe a prototype system and illustrate an example of ball screw design.
, The Japanese Society for Artificial Intelligence, JapaneseSupporting conceptual design based on the function-behavior-state modeler
Y Umeda, M Ishii, M Yoshioka, Y Shimomura, T Tomiyama, AI EDAM-ARTIFICIAL INTELLIGENCE FOR ENGINEERING DESIGN ANALYSIS AND MANUFACTURING, 10, 4, 275, 288, Sep. 1996
The relative significance of conceptual design to basic design or detail design is widely recognized, due to its influential roles in determining the product's fundamental features and development costs. Although there are some general methodologies dealing with functions in design, virtually no commercial CAD systems can support functional design, in particular so-called synthetic phase of design. Supporting the synthetic phase of conceptual design is one of the crucial issues of CAD systems with function modeling capabilities. In this paper, we propose a computer tool called a Function-Behavior-State (FBS) Modeler to support functional design not only in the analytical phase but also in the synthetic phase. To do so, the functional decomposition knowledge and physical features in the knowledge base of the modeler, and a subsystem Qualitative Process Abduction System (QPAS) play crucial roles. Modeling scheme of function in relation with behavior and structure and design process for conceptual design in the FBS Modeler are described. The advantages of the FBS Modeler are demonstrated by presenting two examples; namely, an experiment in which designers used this tool and the design of functionally redundant machines, which is a new design methodology for highly reliable machines, as its application., CAMBRIDGE UNIV PRESS, English設計知識操作論
吉岡真治, 第14回設計シンポジウム講演論文集, 40, 47, 1996知識集約型のエンジニアリング環境の構築(第1報) -基本構成-
石井 理貴, 関谷 貴之, 吉岡 真治, 冨山 哲男, 精密工学会大会学術講演会講演論文集, 1995, 2, 9, 10, 01 Sep. 1995
JapaneseA Design Process Model for Handling Multiple-Design-Object Models.
Yoshioka Masaharu, Nakamura Masahiko, Tomiyama Tetsuo, Yoshikawa Hiroyuki, TRANSACTIONS OF THE JAPAN SOCIETY OF MECHANICAL ENGINEERS Series C, 61, 581, 330, 337, 1995
An intelligent CAD system aims at serving a designer as an intelligent, integrated design environment. For building such a system, development of a large scale knowledge base is necessary. To do so, design knowledge must be systematically formalized, made computable, and organized. Design knowledge has two categories ; i.e., design object knowledge and design process knowledge. In our previous reports, we have formalized a computable design process model to represent design process knowledge and the concept of a metamodel dealing with multiple design object models. This paper proposes a framework to achieve a harmonized combination of these two. In this framework, the computable design process model navigates the metamodel mechanism to generate and modify models of the design object. We describe a prototype system and illustrate an example of ship design., The Japan Society of Mechanical Engineers, JapaneseA Study on an Integrated Modeling Environment for Design (1st Report) : A Proposal of the Pluggable Metamodel Mechanism
YOSHIOKA Masaharu, ISHII Masaki, TOMIYAMA Tetsuo, 人工知能学会全国大会論文集 = Proceedings of the Annual Conference of JSAI, 8, 0, 335, 338, 20 Jun. 1994
JapaneseA Study on an Integrated Modeling Environment for Design (2nd Report) : Integration of a Power Transmission Model
ISHII Masaki, YOSHIOKA Masaharu, TOMIYAMA Tetsuo, 人工知能学会全国大会論文集 = Proceedings of the Annual Conference of JSAI, 8, 0, 339, 342, 20 Jun. 1994
Japanese設計知識表現のためのメタモデルの研究(第12報)
吉岡真治, 1994年度精密工学会春季大会講演論文集, 233, 234, 1994Supporting Design Object Modeling by Metamodel Mechanism.
吉岡真治, 石井理貴, 桐山孝司, 冨山哲男, 人工知能学会知識ベースシステム研究会資料, 26th, 1993Books and other publications
Lectures, oral presentations, etc.
- 知識工学者のための Wikipedia カテゴリの分析とその活用
吉岡真治
人工知能学会 第110回人工知能基本問題研究会, 24 Sep. 2019, Invited oral presentation
[Invited] - Knowledge exploratory project for nanocrystal device development
Masaharu YOSHIOKA
ISMAC 2016 2016 International Symposium on Multimedia and Communication Technology, 02 Sep. 2016, English, Keynote oral presentation
[Invited], [International presentation] - Supporting design and engineering by physical concept ontologies
Masaharu YOSHIOKA
The Tenth International Symposium on Tools and Methods of Competitive Engineering (TMCE 2014), 22 May 2014, English, Nominated symposium
[Invited], [International presentation]
Courses
Affiliated academic society
Research Themes
- Construction and utilization of large scale knowledge based on Wikipedia by considering editorial policy
Grants-in-Aid for Scientific Research
01 Apr. 2024 - 31 Mar. 2027
吉岡 真治, 武田 英明, 大久保 好章
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), Hokkaido University, 24K03042 - Automatic extraction of chemical reaction process information from papers and its utilization
Grants-in-Aid for Scientific Research
30 Jun. 2023 - 31 Mar. 2025
吉岡 真治, 原 真二郎, 長田 裕也
Japan Society for the Promotion of Science, Grant-in-Aid for Challenging Research (Exploratory), Hokkaido University, 23K18500 - 論文内の記述と各種科学技術DBを連携させる特定研究グループ向け論文DBの研究
科学研究費助成事業 挑戦的研究(萌芽)
09 Jul. 2021 - 31 Mar. 2023
吉岡 真治
本研究では、特定の分野に興味を持つ研究者が収集する関係分野の論文を対象としたデータベースに基づいて、専門用語抽出などを行うことにより、用語間の共起関係に基づく分析や、時系列を考慮した研究動向分析を行うデータベースを基礎として、化学物質データベースなどのさまざまな科学技術データベースと連携させることにより、分野の研究者の研究活動を支援するシステムの構築を目指している。本年度は、これまでに作成した論文・図表データベースを様々な科学技術データベースと連携させるための手法について検討すると共に、その機能をこれまでに作成した論文・図表データベースに実装した。本データベースでは、PDFやRSS(Rich Site Summary:論文誌などが論文を発行する際に、abstractなどのメタデータを含む情報を配信する仕組み)を対象として、対象データの収集を行う。
本データベースでは、収集した各論文および、論文中に含まれる図表に対して、本文やキャプションから分類済みの用語抽出を行い、その出現頻度などをメタデータとして付与している。このメタデータを用いた検索を行うことにより、材料としてAを使っている論文(図表)で、利用されることが多いパラメータといった検索が可能となる。また、検索結果を、抽出した専門用語をハイライトすると共に、外部データベースへのリンクが埋め込まれた形で表示することで、科学技術データベースとの連携を実現している。また、論文からの情報抽出の研究として、論文からの化学反応情報の抽出に関する研究を進めると共に、化学反応データベースであるReaxysとの対応関係についての調査を行った。
日本学術振興会, 挑戦的研究(萌芽), 北海道大学, 21K19814 - Measuring Beginner Friendliness / Visual Intelligibility of Web Pages explaining Academic Concepts and Procedures / Know-hows and their Recommendation System
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
01 Apr. 2019 - 31 Mar. 2022
Utsuro Takehito
In this project, we focused on Web pages explaining certain concepts, events, procedures etc. in a beginner friendly manner. In those Web pages, visual intelligibility in Web page layout and beginner friendly Web page texts are the requirements of Web contents for beginners. As a typical case study, this project focused on the factors of Web page visual intelligibility as well as beginner friendliness for explaining academic concepts. Then, this project proposed how to measure those factors as well as to identify those factors. Finally, in this project, we developed a system for recommending those Web pages explaining academic concepts based on those factors.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), University of Tsukuba, 19H04417 - Research on Analysis and Visualization of News Site Stances based on the Comparison of Multiple News Sites
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
01 Apr. 2018 - 31 Mar. 2022
Yoshioka Masaharu
In this research, we proposed a method to analyze news site stance of particular news topic using the distribution of polarity of the articles published by the news site from multiple perspectives. In this research, we confirmed that it is important to understand the interests on subtopics and/or genres of the news articles for detailed analysis of the news site stance. Based on this understanding we proposed a system that can analyze news site stance interactively by taking account the polarity of the related news articles for multiple subtopics and/or genres.
We also proposed Wikipedia category ontology that is new knowledge resource extracted from Wikipedia category from the viewpoint of knowledge engineering. This new knowledge can be used for analyzing the type of the named entity.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), Hokkaido University, 18H03338 - Research on an Paper/Figure Database System with Technical Term Management Ecosystem for Particular Research Group
Grants-in-Aid for Scientific Research
28 Jun. 2019 - 31 Mar. 2021
Yoshioka Masaharu
Objective of this research project is proposal of an paper/figure database system with technical term management ecosystem for particular research group. This system constructs database based on the information (text, figure, caption, and metadata) extracted from the research papers collected by a research group. The system also supports to construct technical term dictionary management system and this dictionary is used for extracting technical terms from the texts. The system also provides a framework to analyze papers using these terms. We implement a system and construct database using research papers of interest and evaluate the effectiveness of the system. We plan to use this database for the ordinal research activities for long term evaluation.
Japan Society for the Promotion of Science, Grant-in-Aid for Challenging Research (Exploratory), Hokkaido University, 19K22888 - Exploratory Search Considering the User's Situation
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (A)
01 Apr. 2016 - 31 Mar. 2020
Kando Noriko
The purpose of this study is to propose the interactive exploratory search technologies supporting users based on each user's situation. Firstly, we have conducted multiple laboratory-typed or crowdsourcing user studies and investigated various factors affect the search process and user's perceived satisfaction, such as the cognitive complexity of the search tasks. Secondly, we have investigated and tested the wide variety of tools and underlying technologies supporting user's exploratory such as inference of the user's situation based on logs and/or eye gaze, query recommendation, search results diversification, multi-facet search, and ostensive search. Thirdly the prototype of the ostensive search model-based interactive exploratory search and guide app for the museum was developed, using user's logs both in the virtual space on the app and the physical space in the museum, encouraged the unforgettable museum experience and introduced the context before and after the museum visits.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (A), National Institute of Informatics, 16H01756 - Revised Algorithm for Maximal Analogies
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)
01 Apr. 2016 - 31 Mar. 2019
Haraguchi Makoto
When we read several stories or legal cases, we often find relational similarities between objects in spite of the difference of stories they belong to. The goal of this research is to design and implement an efficient algorithm for finding those structural similarities among more than two domains. For this purpose, we have designed such an algorithm making use of Non-Negative Matrix Factorization and Graph Mining techniques. The algorithm we have obtained is effective not only for text analysis but also for mining similar structure among databases as data warehouses.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (C), Hokkaido University, 16K00291 - A Study on a System for Comparative Analysis of Web Search Information Needs between Languages
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
01 Apr. 2014 - 31 Mar. 2017
UTSURO Takehito, NAKAGAWA Hiroshi
In this project, we developed a system for comparative analysis of Web search information needs between languages. As for the source of collecting information in multiple languages, we focus on search engine suggests as well as question-answer sites in multiple languages. Given a specific query focus, we collect search information needs as well as Web pages and question-answer pairs for those search information needs. Then, developed a system for assisting the process of comparatively analyzing those collected search information needs and the specified Web pages / question-answer pairs between multiple languages and that of discovering differences of culture, concern, and opinions between multiple countries.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), University of Tsukuba, 26280111 - Information Extraction from Nanocrystal device development papers and its utilization
Grants-in-Aid for Scientific Research Grant-in-Aid for Challenging Exploratory Research
01 Apr. 2014 - 31 Mar. 2016
Yoshioka Masaharu, HARA Shinjiroh
In this project, we have constructed a corpus for extracting useful information related to the experiment of nanocrystal device development (e.g., Source material and parameters) and developed an automatic information extraction system NaDev that tries to extract such information from papers. By using NaDev, we can extract characteristic keywords related to the experiment (e.g., source material, experimental parameter) and these extracted keywords are used for calculating similarity among different papers and information retrieval system that have query construction support. In this project, we proposed a figure retrieval system that uses such parameter information as query.
Japan Society for the Promotion of Science, Grant-in-Aid for Challenging Exploratory Research, Hokkaido University, Principal investigator, Competitive research funding, 26540165 - Named entity recognition using Linked Open Data and its application to information retrieval
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
01 Apr. 2013 - 31 Mar. 2016
Yoshioka Masaharu, KANDO Noriko
In this research, we proposed a framework to extract named entity from the text with named entity type information by using DBPedia. We also proposed facet-biased topic model that take into account the type of named entity as facet. In addition, we implemented WC3 that analyzes consistency of metadata in articles that belongs to a particular Wikipedia category. We confirmed that WC3 can find out the articles that have inconsistent metadata and candidate articles for the category.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), Hokkaido University, Principal investigator, Competitive research funding, 25280035 - A System for Multi-Faceted Topic Analysis and Cross-Lingual Comparative Web Text Analysis
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
01 Apr. 2011 - 31 Mar. 2014
UTSURO TAKEHITO, YOSHIOKA Masaharu, INUI Takashi, NAKAGAWA Hiroshi, KIYOTA Yoji
In this project, given a collection of multilingual documents available on the Web, we cross-lingually and cross-culturally compare facts and opinions that are observed in the collected documents. We especially focus on various facets including the following four of (i) to (iv), and realize a framework of assisting the process of discovering cross-cultural differences: (i) differences in specific issues and concerns within a topic, (ii) differences in time series features, (iii) differences in real life experiences, (iv) differences in opinions.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), University of Tsukuba, Coinvestigator not use grants, Competitive research funding, 23300033 - Integrated analysis of web information structure and users' behavior and its application to advanced information access
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (A)
01 Apr. 2010 - 31 Mar. 2013
OYAMA Keizo, AIZAWA Akiko, MIYAO Yusuke, SUN Yuan, KOBAYASHI Tetsuro, HAN Hao, KISHIDA Kazuaki, YAMANA Hayato, OKUMURA Manabu, YOSHIOKA Masaharu, ISHITA Emi, MURATA Tsuyoshi, EGUCHI Koji
For understanding Web structure and users' behavior of information retrieval and browsing in an integrated way, and for extending it to various applications, we collected and introduced various data reflecting Web information structure and Web users' behavior (e.g. Web view log data, micro-blog data), obtained user data through questionnaire, and executed integrated analysis on them.
Consequently, we obtained various findings through data such that there is a gap between information wanted to know and information wanted to inform, and that, through using Web portal sites, unexpected contact to various information occurs. Moreover, we proposed and studied various methods for advanced information access such as information recommendation and information retrieval based on the information obtained through the integrated analysis.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (A), National Institute of Informatics, 22240007 - Research on world news analysis system that supports to compare news site with different characteristics
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
2009 - 2011
YOSHIOKA Masaharu, KANDO Noriko, SEKI Yohei
In this research, we propose a news analysis system that supports to compare articles from news sites in different countries. This system can support to analyze articles from multi facet (e. g., opinion, named entity). In this research, we propose to construct named entity database and opinion analysis system. In addition, we also propose a framework to evaluate such interactive system by using information that characterizes change of the users' knowledge about task domain.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), Hokkaido University, Principal investigator, Competitive research funding, 21300029 - 個々のWebサイトが持つ情報の特性を考慮した多角的意見集約・分析の研究
科学研究費助成事業 特定領域研究
2007 - 2008
吉岡 真治, 原口 誠, 大久保 好章
本研究では、情報抽出元である各々のWebサイトが持つ情報の偏りなどの特徴を分析する方法を提案すると共に、この特徴分析の結果を踏まえた意見集約・分析の方法を提案している。具体的には、全てのデータセットを含む全体のデータベースと特定の条件を満たす部分データベースにおけるデータの分布の違いに注目したコントラストセットマイニングを適用することにより、個々のサイトの特徴を分析する方法を提案する。
この目標に対して、本年度は、「コントラストセットマイニングによる個々のサイトの特徴分析」の考え方に基づくシステムを実装し、様々なレベルのユーザ実験を通した評価を行った。簡単にシステムを使用してもらってコメントをもらうというユーザ実験では、半数以上の利用者から、システムが有用であるというコメントをもらえた。一方、レポート作成課題の実施という本システムの利用目的に応じた課題の遂行という実験を行ったところ、異なる国のデータとの比較を行う際には、外国語の新聞を直接の比較対象とする必要性などが指摘された。この指摘を受け、機械翻訳システムと組み合わせることによりシステムを作成した。このシステムでは、ニュースなどの時事的なトピックに対する対訳辞書の不備を補うためのWikipediaを用いた対訳辞書の作成方法などの提案も行った。また、ピンポイントクラスタリングによるWebサイトの分析」の課題については、昨年度から継続して、フォーマルコンセプト(FCA)を用いた分枝限定法にもとづくトップN-FC探索手法の研究を行った。
日本学術振興会, 特定領域研究, 北海道大学, Principal investigator, Competitive research funding, 19024002 - Crossover Concept Mining Based on Conceptual Abstraction Hierarchies
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
2007 - 2008
MAKOTO Haraguchi, ETSUJI Tomita, KIMIHITO Ito, MASAHARU Yoshioka, YOSHIAKI Okubo
潜在的なクラスタ間接続を, たとえ, 外延的な重なりが小さな場合でも, 内包的な繋がりが高い場合は有意義なクラスタ間接続関係を与える「情間」と捉える.こうした「情間」は, クラスタの外延と内包に対する評価関数を設定することにより, 制約条件下の最適化問題として捉えることができ, 具体的には, 形式概念束探索における分枝限定アルゴリズムとして設計・開発・実証を行った.形式概念束は, 概念の抽象階層と見なすことができ, 主要なクラスタを繋ぐ適切な概念を抽象階層から見出すことを意味している.実証実験において, カテゴリとしては離れた異種文書群から, それらを跨ぐ, 機能的な概念を情間(クロスオーバ概念)として抽出することに成功している.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), Hokkaido University, Coinvestigator not use grants, Competitive research funding, 19300040 - 問題解決を目的とした文書作成支援型知識マネージメントシステムの研究
科学研究費助成事業 若手研究(B)
2005 - 2007
吉岡 真治
本研究では、問題解決を目的とした文書作成支援型知識マネージメントシステム(Documentation MAnagementsystem for Problem Solving:DMaPS)の構築を目的とした研究を実施した。このシステムでは、問題解決の支援、文書と形式化された知識の蓄積、文書からの知識の獲得というループの実現を目指している。この目的に対し、平成17年度は、問題解決の支援を行うために、ユーザが持っている問題に関する記述と類似している文書知識を提示することにより、問題解決の支援を行うシステムを作成した。本システムに対し、あらかじめ手作業により作成した知識を埋め込んだ文書群を与えることにより、問題解決の支援が可能であることを確認した。平成18年度以降は、主に、一般的な文書から問題解決に役立つ知識を獲得するための方法論に注目して研究を遂行した。具体的には、トラブル事例に関する知識獲得を目指して、新聞記事などから、ユーザが直面している問題に関連する訴訟の記事などを検索し、その検索結果から、関係する利害関係者とその関係を抽出するシステムの構築を行った。本システムでは、少数の訓練事例にたいして、複数の基準を用いた機械学習システムを適用し、その結果を組み合わせることによって、人手で作成したパターンでは抽出することができなかった多くの利害関係者の情報を獲得することが確認できた。また、文書分類の研究では、特許を関連する物理法則によって分類する考え方を提案し、Web上に存在する様々な物理法則に関する文書の情報などを利用して、具体的な特許文書を分類する方法を提案した。ただ、これらで得た知識は、DMaPSで利用可能なレベルの知識での形式化が困難であったため、今後は、形式的な知識作成に関する検討を進める必要がある。
日本学術振興会, 若手研究(B), 北海道大学, Principal investigator, Competitive research funding, 17700087 - 適合的汎化に基づく情報検索システムの実験的研究
科学研究費助成事業 特定領域研究
2004 - 2005
吉岡 真治, 原口 誠, 大久保 好章
本研究では、検索意図を推定し、それをユーザにも理解しやすい形式で表現できる情報検索システムを提案する。ここでは、検索者が用いる検索語の抽象度に注目して、検索意図の推定を行う。しかし、ユーザは、検索意図を表現するのに適切な抽象度の概念を必ずしも用いない場合がある。そのため、検索意図に応じた適切な抽象度の概念を検索キーワードに用いると、ユーザにも理解しやすく効率的な検索キーワードになると考えられる。
そのため、本研究では、電子化辞書やシソーラスに記述されている概念階層構造を利用し、検索意図に応じて検索キーワードを適合的に汎化し、新たな検索式を作成する情報検索システムを提案する。ここで、適切な抽象度の汎化とは、検索キーワードが持つ正解判定の分別能力に関する情報を多く保存する汎化の事である。
本年度は、検索式における検索語と検索要求のミスマッチという観点から検索課題の持つ性質の分類を行うと共に、検索語拡張に与える影響を分析した。その結果、検索語と検索要求のミスマッチのタイプに応じて、検索語拡張に求められる機能も異なり、単一の検索語拡張手法では、検索性能を改善させる場合だけでなく、性能を悪化させる場合があることが確認された。よって、検索語拡張の洗練化を行うためには、検索語と検索要求のミスマッチに応じた検索語拡張の方法の提案と、ミスマッチのタイプの推定を行うことが必須であることを提案した。
日本学術振興会, 特定領域研究, 北海道大学, Principal investigator, Competitive research funding, 16016201 - A Text Organization Method Based on Maximal Analogy
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
2004 - 2005
HARAGUCHI Makoto, TANAKA Yuzuru, KAKUTA Tokuyasu, YOSHIOKA Masaharu, OKUBO Yoshiaki
In the research project, an algorithm for extracting common abstract event sequences, given two or more documents, is presented. In order to avoid combinatorial explosions in matching more than two documents, the algorithm consists of two phases.
The first phase is basically a text summarization system taking balances between the importance of sentences in each segment in a given document and the importance of sentences to connect several segments. The latter importance is used to extract contextual sentences involving contextual words. In order to separate the notion of importance into the two as in the above, we compute a chunk of core sentences in each segment by a clique finding algorithm, and then calculate the degree of latter importance of sentences from the chunk just in the way used in KeyGraph. Finally, the overall importance is determined by a scheme very similar to topic-sensitive PageRank. We have made some experiments for newspaper articles, and verified its effectiveness.
In the second phase, we use the summarized document obtained in the first phase as a kind of source document. The summarized document preserves the structure of the original document at more abstract level. Therefore, for each event in the source, it suffices to find a similar event in a given target document. This drastically reduces the computational complexity need to build correspondence between the two documents. As a result, several document set is now summarized from a viewpoint of the source document by means of analogy.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), Hokkaido University, Coinvestigator not use grants, Competitive research funding, 16300039 - Research on design support by creative abduction
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
2002 - 2004
TAKEDA Hideaki, SHIMOMURA Yoshiki, YOSHIOKA Masaharu
This project presents "Universal Abduction Studio" (UAS) that is a new approach to computational support for designers and discusses technologies to implement it. Although design support systems have been well developed for geometric and detail design stages, they are still unsuccessful in the conceptual design stage. In our opinion, the main difficulty comes from incomplete and insufficient understanding about design knowledge and its operations that play a crucial role in conceptual design. In contrast, however, recently thanks to development of the Internet technologies, more and more knowledge is accumulated and available electronically. It then becomes an interesting research question how to apply such an enormous amount of diverse knowledge to conceptual design. UAS is not intended to be an automated design system. UAS is an environment in which designers can manage and apply knowledge to arrive at new, creative design collaboratively with the system. UAS provides the designer with knowledge and reasoning functions, while the designer selects operations and evaluates design results iteratively. we show a novel knowledge representation to support creative design process. Our RDF-based representation consists of modules for knowledge repositories like predicate and object ontologies. A user can compose an individual model for knowledge by selecting and combining suitable elements in these knowledge repositories. This representation solves two characteristics observed in our case study analysis, ie., multiplicity of interpretation and variety of knowledge. The graph structure can generate multiple interpretations with minimum modification such as redirecting "if" and "then" links. Furthermore, since we separate metamodel and aspect-specific models, we can obtain various aspect-specific models from a single metamodel. The problem to handle with variety of knowledge is balance of expansivity and consistency. Since we provide categories of knowledge like object and predicate ontologies in order to clarify roles of knowledge, we can add easily new knowledge like domain-specific knowledge according to its roles without avoiding messy.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), National institute of Informatics, Coinvestigator not use grants, Competitive research funding, 14380170 - 適合的汎化に基づく情報検索システムの実験的研究
科学研究費助成事業 特定領域研究
2003 - 2003
吉岡 真治, 大久保 好章, 原口 誠
本研究では、検索意図を推定し、それをユーザにも理解しやすい形式で表現できる情報検索システムを提案する。ここでは、検索者が用いる検索語の抽象度に注目して、検索意図の推定を行う。しかし、ユーザは、検索意図を表現するのに適切な抽象度の概念を必ずしも用いない場合がある。そのため、検索意図に応じた適切な抽象度の概念を検索キーワードに用いると、ユーザにも理解しやすく効率的な検索キーワードになると考えられる。
そのため、本研究では、電子化辞書やシソーラスに記述されている概念階層構造を利用し、検索意図に応じて検索キーワードを適合的に汎化し、新たな検索式を作成する情報検索システムを提案する。ここで、適切な抽象度の汎化とは、検索キーワードが持つ正解判定の分別能力に関する情報を多く保存する汎化の事である。
本年度は、この適合的汎化による検索拡張が、正解文書において補完関係にある検索語の発見を目指しているという点に注目し、適合的汎化による検索拡張の結果をBoolean式として表現する方法について検討を行った。さらに、実際にユーザが利用可能なインターフェースの作成を行った。また、文書中の語の役割を反映した抽象概念の提案手法についても検討を行った。これらの手法の有効性は、実際のユーザ実験によって検討する必要性がある。次年度以降は、ユーザ実験などを通して、本手法の有効性の検証と手法の洗練化を図っていきたいと考えている。
日本学術振興会, 特定領域研究, 北海道大学, Principal investigator, Competitive research funding, 15017202 - 適合的汎化に基づく情報検索システムの実験的研究
科学研究費助成事業 特定領域研究
2002 - 2002
吉岡 真治, 大久保 好章, 原口 誠
本研究では、検索者にも理解しやすい概念階層の汎化という考え方を利用して、ユーザの検索意図を明示化すると共に、精度落ちを抑えた情報検索システムを提案している。この概念階層を利用する情報検索システムでは、汎化の対象としてどの様な語を選ぶのかという問題に加え、あらかじめ用意する辞書における概念階層の性質に結果が依存するという問題がある。
本年度は、汎化のための方法論として、適合文書群に共通する概念を利用する方法を検討した。学術論文のアブストラクトを用いたNTCIRや新聞記事を用いたIREXの2つの情報検索のテストコレクションを利用し提案手法のシステムの検索性能への影響を検討した。その結果、提案手法により検索性能が向上することが確認できた。
次に、辞書の性質を議論するために、昨年度までの研究で用いていたEDR電子化辞書に加え、日本語概念語彙体系という異なるタイプのシソーラスを用いた場合の影響に関する分析を行った。検索実験の結果、日本語概念語彙体系を用いた場合に、与えた適合文書へのオーバーフィットの度合いが高いことが確認された。これは、日本語概念語彙体系とEDRが持つ概念カテゴリーの粒度の違いによるものであると考えられる。日本語概念語彙体系とEDRが持つ概念カテゴリーの数を単純に比較すると、EDRは約50000個(専門用語辞書のものを含む)であり、日本語概念語彙体系は約3000である。よって、適合的汎化による情報検索システムで用いるシソーラスは、ある程度、粒度の細かい概念を扱うことができるシソーラスであることが望まれることが確認された。
日本学術振興会, 特定領域研究, 北海道大学, Principal investigator, Competitive research funding, 14019002 - 設計対象モデルを用いた設計文書の体系的管理と再利用に関する研究
科学研究費助成事業 若手研究(B)
2001 - 2002
吉岡 真治
本研究では、設計問題に代表される問題解決の過程を問題を表現するためのモデルとそこで用いられた知識さらに、その際に記述された文書を組み合わせて保存することにより、設計文書の体系的管理と再利用を実現することを目的とした。その成果として、問題解決のための文書管理システムDMaPS(Documentation Management system for Problem Solving)を提案した。
このDMaPSでは、ひとつのサブ問題の問題解決の過程を単位として考えた文書知識を利用して、文書の管理と問題解決の支援を行う。この文書知識では、文書を単純なテキスト情報として保存するだけではなく、知識を適用する背景となる条件や形式的な知識を一階述語論理によって表現する、この二つの表現を組み合わせることにより、マニュアルなどの形式知とノウハウのように形式化が困難な情報を組み合わせて取り扱うことが可能となる。
本システムでは、この文書知識を用いることにより、ユーザの問題解決の支援を行う。具体的には、ユーザが直面している問題との一階述語論理による形式的表現に関する類似性とキーワードによる情報検索による類似性という異なる観点から知識の検索を行うことができ、より柔軟な知識の検索を実現する。また、検索された知識を取り込むことにより、問題解決の過程を説明するための文書が半自動的に作成でき、ユーザは個別事例に適応させる部分の記述に専念することができる。さらに、本システムでは、知識のバージョン管理の機構を取り込んでおり、各文書問題解決の際に利用した知識のバージョンをチェックすることにより、既存の設計文書をそのまま再利用してよいのか、再検討が必要なのかを調べることが可能になっている。
本研究では、建築構造物の基礎設計を例にとり、知識記述の実験を行い、体系的な知識整理が可能であることを示した。
日本学術振興会, 若手研究(B), 北海道大学, Principal investigator, Competitive research funding, 13750129 - 適合的汎化に基づく情報検索システムの実験的研究
科学研究費助成事業 特定領域研究(C)
2001 - 2001
吉岡 真治, 大久保 好章, 原口 誠
本研究では、検索意図を推定し、それをユーザにも理解しやすい形式で表現できる情報検索システムを提案する。ここでは、検索者が用いる検索語の抽象度に注目して、検索意図の推定を行う。
しかし、ユーザは、検索意図を表現するのに適切な抽象度の概念を必ずしも用いない場合がある。そのため、検索意図に応じた適切な抽象度の概念を検索キーワードに用いると、ユーザにも理解しやすく効率的な検索キーワードになると考えられる。
そのため、本研究では、電子化辞書やシソーラスに記述されている概念階層構造を利用し、検索意図に応じて検索キーワードを適合的に汎化する情報検索システムを提案する。ここで、適切な抽象度の汎化とは、検索キーワードが持つ正解判定の分別能力に関する情報を多く保存する汎化の事である。
一方、電子化辞書やシソーラスでは、一般的な概念階層が記述されているために、多段階の階層を用いた汎化を行った際に、一度に大量の語を代表する概念に汎化されてしまう問題がある。例えば、「ビデオ」と「DVD」を汎化した概念として「映像装置」という概念を考えたときに、8mmカメラは含むことは、検索意図と必ずしも一致しないと考えられる。そのため、ユーザが持つ細かな検索意図を適切に表現するために、検索目的に応じた適切な抽象概念を設定し、概念階層を再構築することが必要である。
本年度は、概念の適切な抽象化のための指標として、キーワードと正解文書の相互情報量に基づく関連性の指標を利用し、情報検索システムを構築した。本システムは、5件の正解文書を与えることにより、検索語の汎化を行い、検索者の意図が理解しやすい検索式が作成可能である。また、この検索式による検索は、他の検索システムと比較して十分な検索精度を持っていることを確認した。
日本学術振興会, 特定領域研究(C), 北海道大学, Principal investigator, Competitive research funding, 13224001 - A stydy knowledge lifecycle for commmmumication support on the netwook
Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
1999 - 2001
TAKEDA Hideaki, TAKEUCHI Koichi, UENO Atsushi, NISHIDA Toyoaki
In this research, we proposed concept "knowledge as media" to realize knowledge lifecycle for communication support on the network. In the traditional view of artificial intelligence, Knowledge exists within agents such as human and computer programs. On the contrary, we regard knowledge exists among agents, i. e., knowledge is media which enables communication among agents. Under this view, we investigated knowledge from the abstract way to the practical one. Ontology is the most abstract form of knowledge, knowledge for community and knowledge for environments are intermediate, and the embodied knowledge is the most practical one. We have three major results. One is community-support system called TelMeA in which interface agents as participants can form a field of communication. We tested this system by the psychological experiment and showed that using interface agents like this system is effective for communication. The second result is a system to capture personal knowledge. The system called MindHeap can gather knowledge through WWWW browsing. If a user can pick up sentences in WWWW, the system memorizes the sentences and their contexts. The sentences are connected to each other through their contexts in this system. The user can retrieve related memorized sentences automatically and even summarize WWW pages based on memorized information. The third result is embodiment-based learning, in which embodiment knowledge is captured through actions. The agent can learn how its body is by moving around and sensing the environment.
Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), 奈良先端科学技術大学院大学->国立情報学研究所, Coinvestigator not use grants, Competitive research funding, 11480078 - Studies on International Collaboration for Research Networking and International Coordination of Asia-Pacific Coutries
Grants-in-Aid for Scientific Research Grant-in-Aid for international Scientific Research
1995 - 1997
INOSE Hiroshi, SAKAI Yoshinori, SEZAKI Kaoru, IIDA Noriko, YASUDA Yasuhiko, FUJINO Takayuki, YOSHIOKA Shinji, ZHAO Ihei, ABE Shuji, UOSE Hisao, LANDWEBER Lawrence, AIZAWA Akiko, KEI Usei, OHYAMA Keizo, MATSUKATA Jun, HASHIZUME Hiromichi, ADACHI Jun, MIYAZAWA Akira, ASANO Shoichiro, HAMADA Takashi, ONO Kinji
In recent years, research networks have extremely diversified, not only in developed coutries but also in newly developing contries and region, including mid and east europe, mid and south america, africa and asian coutries. As the result, international coordination among research networks is of important to establish international communication links. The CCIRN (Coordination Committee for Intercontinental Research Networks) plays primary role of such coordination. In addition to this, need for braodband communiation links is emphasized in order ato stimulate higher end researches. In this context, the CCIRN is now discussing how to interwork such broadband communication links and networks.
United States had lauched new interconnecting point among broadband networks in Chikago, IL., which is called STAR-TAP (Science, Technology and Research, Transit Access Point). Investigators joined to the study have made international cooperation activities during three years, and finally made success to interconnected japanese boradbnad network (SINET) to STAR-TAP in October, 1997.
European Commuinity has also put in operation of braodband research network (TEM-34). SINET has been cooperating to european resarch networks for three years, and consequently SINET has changed its connection point to TEM-34 PoP (Point of Presence) to London, UK.Circuit to London will be in operation from early April, 1998. That is second result of the study.
Beside the actual interconnection, investigators have been contributing international standardization at IETF on ATM based networking. Several proposals which are now available as Internet Drafts, including ATM address resolution scheme, ATM inteconnection control information and high-performance protocol sepcifications, are proposed from the study.
These activities are evaluated as future basis for international connection of research networks.
Japan Society for the Promotion of Science, Grant-in-Aid for international Scientific Research, 学術情報センター, Coinvestigator not use grants, Competitive research funding, 07041116 - 大規模知識ベースを用いた設計支援の研究
1996
Competitive research funding
Educational Organization
- Bachelor's degree program, School of Engineering
- Master's degree program, Graduate School of Information Science and Technology
- Doctoral (PhD) degree program, Graduate School of Information Science and Technology