SEARCH

Search Details

Minami Hiroyuki

Information Initiative Center　Cyber SecurityProfessor
ICT Promotion OfficeProfessor
Institute for the Promotion of Business-Regional CollaborationProfessor

Researcher basic information

■ Degree
  • Ph.D (Eng.), Hokkaido University, Mar. 1994
  • Master Degree, Hokkaido University, Mar. 1991
  • 文学士, 北海道大学, Mar. 1989
■ URL
researchmap URL■ Research Keywords and Fields
Research Keyword
  • Symbolic Data Analysis
  • Cyber Security
  • データ解析
  • データ解析システム
  • インターネット
  • 知識工学
  • エキスパートシステム
  • Internet
  • 統計ソフトウェア
  • 仮想並列計算機環境
  • Object-Oriented
  • GUI
  • 並列データ解析
  • ブートストラップ法
  • リサンプリング
  • 統計パッケージ
  • 総合評価
  • 関数クラスター分析
  • 関数データ
  • トラヒック
  • 統計グラフ
  • 高速計算
  • コンピュータネットワーク
  • 計算機統計学
  • Computer Network
  • Computational Statistics
Research Field
  • Informatics, Information security
  • Informatics, Information network
  • Informatics, Statistical science
  • Informatics, Computer system
■ Educational Organization

Career

■ Career
Career
  • Oct. 2015 - Present
    Hokkaido University, Information Initiative Center, Professor
  • Apr. 2007 - Sep. 2015
    Information Initiative Center, Hokkaido University, Associate Professor
  • Apr. 2003 - Mar. 2007
    Information Initiative Center, Hokkaido University, Associate Professor
  • Nov. 1999 - Mar. 2003
    Center for Information and Multimedia Science, Hokkaido University, Associate Professor
  • Apr. 1994 - Mar. 2000
    Information and Management Science, Faculty of Commerce, Otaru University of Commerce, Associate Professor
Educational Background
  • Apr. 1991 - Mar. 1994, Hokkaido University, Graduate School of Engineering, Doctoral Course, Division of Information Engineering, Japan
  • Apr. 1989 - Mar. 1991, Hokkaido University, Graduate School of Engineering, Master Course, Division of Information Engineering, Japan
  • Apr. 1985 - Mar. 1989, Hokkaido University, Faculty of Letters, Behavioral Science, Japan
Committee Memberships
  • Apr. 2023 - Present
    札幌市情報公開・個人情報保護審議会, 委員, Autonomy
  • Jan. 2023 - Present
    日本計算機統計学会, 理事, Society
  • Apr. 2021 - Present
    国立情報学研究所 学術情報ネットワーク運営・連携本部, 委員, Others
  • 2015 - Present
    国立情報学研究所 学術情報ネットワーク運営・連携本部, セキュリティ運営委員会 委員, Others
  • 1994 - Present
    日本計算機統計学会, 学会誌編集委員（和文誌）, Society
  • Jan. 2019 - Dec. 2022
    日本計算機統計学会, 監事, Society
  • Jun. 2017 - May 2021
    統計数理研究所, 共同利用委員会 委員, Others
  • 1994 - Dec. 2018
    日本計算機統計学会, 理事, Society
  • Jan. 2011 - Dec. 2012
    日本計算機統計学会, 評議員, Society
  • Apr. 2002 - Mar. 2003
    札幌市, IT経営戦略アドバイザリースタッフ, Autonomy

Research activity information

■ Awards
■ Papers
  • A Study on Anomaly Detection Methods Using ICSOnt Ontologies for Industrial Control Systems in Factories
    水谷駿斗; 近藤賢郎; 石原真太郎; 前田一平; 金野勇樹; 田中里咲; 奥村 剛; 南 弘征
    信学技報, 125, 405, 16, 23, 16 Mar. 2026, [Last author]
    Japanese, Symposium
  • GAMPALv2.5: Enhancing Large-scale Scanning Attack Inference in General-purpose Internet Traffic Anomaly Detection Mechanism
    Keisuke Tachibana; Takao Kondo; Hiroyuki Minami
    2026 International Conference on Computing, Networking and Communications (ICNC), 484, 490, IEEE, 16 Feb. 2026, [Peer-reviewed], [Last author]
    English, International conference proceedings
  • Early Anomaly Detection Mechanism in Darknet Traffic focusing on Communication Behavior from Residential IP addresses
    古山輝龍; 近藤賢郎; 南 弘征
    信学技報, 125, 320, 150, 155, 23 Jan. 2026, [Last author]
    Japanese, Symposium
  • 長打傾向に基づく MLB選手の打撃タイプ分類法に関する一考察
    松本 大輝; 近藤 賢郎; 南 弘征
    2025年度 日本分類学会 シンポジウム 予稿集, 39, 42, Dec. 2025, [Last author]
    Japanese
  • 眼底写真の血管形態定量化処理に依る特徴量と 健診データを用いた多角的解析
    大坪周平; 福津佳苗; 齋藤理幸; 中野義雄; 南弘征; 水田正弘
    2025年度日本分類学会 シンポジウム予稿集, 9, 12, Dec. 2025
    Japanese, Symposium
  • DIAMEC：Multi-Domain Infrastructure of Authentication and Authorization on the Wide Area Edge-Computing Environment
    杉田翔梧; 近藤賢郎; 南 弘征
    IPSJ SIG Technical Report, 70, 2, Jul. 2025
    Japanese, Symposium
  • インターネットトラフィック汎用異常検知手法GAMPALにおける大規模スキャニング攻撃の検知性能向上
    橘 慶祐; 近藤 賢郎; 南 弘征
    信学技報, 125, 85, 78, 85, 24 Jun. 2025
    Japanese, Symposium, 48594685
  • レシートデータの商品名表記揺れの補正とトピックモデルに基づくチェーン別傾向分析
    杉田翔梧; 橘 慶祐; 松本大輝; 大坪周平; 近藤賢郎; 南 弘征; 水田正弘
    日本計算機統計学会第39回大会 講演報告集, 114, 117, 13 Jun. 2025
    Japanese, Symposium
  • Cyber-Attack Investigation with Symbolic Data Analysis and its practical application
    Hiroyuki MINAMI; Masahiro MIZUTA
    IASC-ARS Interrim Conference 2024, Dec. 2024, [Peer-reviewed], [Lead author, Corresponding author]
    English, International conference proceedings
  • Attribution Method for Early Detection of Attack Activities in Darknet Traffic
    Kiryu, Koyama; Takao, Kondo; Hiroyuki, Minami
    コンピュータセキュリティシンポジウム2024論文集, 1609, 1616, 情報処理学会, 15 Oct. 2024
    Japanese, 近年，サイバー空間上で観測される攻撃活動は増加の一途にあり，それらの脅威による被害の拡大を防ぐためには攻撃活動の早期検知が重要である．このため，インターネットアプリケーションにおけるソフトウェア脆弱性に対する攻撃者のスキャン活動の早期検知などを目的として，本来割り当てられていない IP アドレス範囲（ダークネット）で観測されるトラフィックの解析が行われている．通常，ダークネットトラフィックでは様々な意図をもった攻撃者による偵察活動や定常的なノイズトラフィックが混在して観測されるほか，攻撃者が発信元を隠ぺいするためにプロキシサービスを用いた攻撃トラフィックも含まれるなど，どのソフトウェア脆弱性を対象としているかなどの属性付けは一般に困難である．本研究では，TCP によるダークネットトラフィックを対象として，プロキシサービスを用いないスキャン活動を推定し，攻撃活動の早期検知に資する属性付けが可能となるかを検証する．まず，GeoIP 情報に基づいてダークネットトラフィックの送信元 IP アドレスを国・地域毎に集約することにより，ダークネットトラフィックを分類する．その後，国・地域毎に観測されるフロー数を集計してスペクトル解析を行い，プロキシサービスと推定できる IP アドレスの割合が少ない群を特定する．次に，特定された群に対して，ARIMA モデルによるフロー数の予測値と実測値の比較に基づいた異常検知機構を構築し，MITRE CVE により脆弱性が公知となった日時よりもどの程度早期の検知が可能となるかを検討する．
    Cyber-attack activity has recently accelerated, and we realize the importance of the early detection to prevent us from damage and its spread. An analysis of Darknet traffic, observed in unallocated IP address ranges, would detect scanning activities against software vulnerabilities in the Internet applications. The traffic consists of various activities, not only dark surveillance but also just background noise. Thus, we can hardly classify attribution of attacks (e.g. for one specific vulnerability). In this study, we discuss the classification about TCP darknet traffic toward early detection of attack activities with estimating scanning activities (excluded via proxy services). We start to classify darknet traffic based on GeoIP information into countries and regions. After that, we perform spectral analysis of flow counts, to identify a group with fewer proxy-associated IP addresses. We implement an anomaly detection mechanism to compare estimates on flows with ARIMA model with the measured ones. Applying it to the groups, we confirm the effectiveness for earlier detection of vulnerabilities than MITRE CVE disclosure.
  • 関数データ解析による飲料製品の購買動向推移に関する考察
    藤嶋 海斗; 大坪 周平; 古山 輝龍; 近藤 賢郎; 南 弘征; 水田 正弘
    日本計算機統計学会シンポジウム論文集, 38, 160, 163, 東京 : 日本計算機統計学会, Oct. 2024
    Japanese
  • 飲料商品POSデータに基づいた販売実績に係る分析と解釈について
    大坪 周平; 古山 輝龍; 藤嶋 海斗; 近藤 賢郎; 南 弘征; 水田 正弘
    日本計算機統計学会シンポジウム論文集, 38, 164, 167, 東京 : 日本計算機統計学会, Oct. 2024
    Japanese
  • Security Information and Event Management (SIEM) application with Symbolic Data Analysis and its implementation
    Hiroyuki MINAMI
    The 12th conference of the Asian Regional Section of the International Association for Statistical Computing (IASC-ARS), 07 Dec. 2023, [Peer-reviewed], [Invited]
    English, 42162895
  • A method for extracting nonlinear structure based on measures of dependence
    Shoma Ishimoto; Hiroyuki Minami; Masahiro Mizuta
    Japanese Journal of Statistics and Data Science, Springer Science and Business Media LLC, 01 Sep. 2022, [Peer-reviewed]
    English, Scientific journal, 33977562
  • An empirical study on nonlinear structure extraction with measures of dependence
    Shoma Ishimoto; Hiroyuki Minami; Masahiro Mizuta
    Book of Abstracts, 24th International Conference on Computational Statistics (COMPSTAT 2022), 50, 25 Aug. 2022, [Peer-reviewed]
    English, International conference proceedings, 33977562
  • A method for extracting nonlinear structure based on nonlinear measures of dependence
    Shouma Ishimoto; Hiroyuki Minami; Masahiro Mizuta
    Book of Abstracts, The 11th Conference of the IASC-ARS (The Asian Regional Section of the International Association for Statistical Computing), 26, Feb. 2022
    English, International conference proceedings
  • Hybrid adaptive index model for binary response data
    Ke Wan; Kensuke Tanioka; Hiroyuki Minami; Masahiro Mizuta; Toshio Shimokawa
    Japanese Journal of Statistics and Data Science, 4, 1, 299, 315, Springer Science and Business Media LLC, Jul. 2021, [Peer-reviewed]
    English, Scientific journal
  • Application of Adaptive Index Model to Evaluate Streetscape Convenience Evaluation Survey Data in Chengdu
    WAN Ke; YAN Kun; LI Li; MINAMI Hiroyuki; SHIMOKAWA Toshio; MIZUTA Masahiro
    Transactions of Japan Society of Kansei Engineering, 20, 1, 27, 37, Japan Society of Kansei Engineering, 2021, [Peer-reviewed]
    Japanese, Scientific journal,

    China has become the world's second largest economy after recent rapid economic growth. On the other hand, the urbanization of China has aggravating the problems of air pollution, traffic congestion, and the dual structure of urban and rural areas. Therefore, the Chinese government has implementing a policy to revise the efficiency-oriented city planning to the quality-oriented, namely "National new urbanization plan". We conducted a questionnaire survey on streetscapes in Chengdu, China. In this survey, we asked pedestrians and bicycle users about the streetscape during traffic. We applied AIM (Adaptive Index Model) method [1] to analyze this survey data. We used the estimated AIM results to score the satisfaction of the streetscape based on "production rules". As a result, safety was important for each traffic, and street cleanliness and the number of bicycle parking lot was important for cyclists. Moreover, Satisfaction was low in the school district and the vicinity of parks. On the other hand, the streets were highly satisfied in the shopping district.

  • Multivariate Analysis for Molecular Species of Cholesteryl Ester in the Human Serum
    Yifan Chen; Shu Ping Hui; Yusuke Miura; Sota Kato; Toshihiro Sakurai; Zhen Chen; Emiko Okada; Shigekazu Ukawa; Takafumi Nakagawa; Koshi Nakamura; Akiko Tamakoshi; Hitoshi Chiba; Hiroyuki Minami; Masahiro Mizuta
    ANALYTICAL SCIENCES, 36, 3, 373, 378, Mar. 2020, [Peer-reviewed]
    English, Scientific journal
  • Functional clustering approach for analysis of concentration
    Yifan Chen; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Studies in Classification, Data Analysis, and Knowledge Organization, 187, 198, 2020
    International conference proceedings
  • Statistical Analysis of Unauthorized-Access Log Data and Its Interpretation
    Hiroyuki MINAMI
    Archives of Data Science, Series A, 6, 1, 13, 22, 2020, [Peer-reviewed], [Lead author, Corresponding author], [International Magazine]
    English, Scientific journal
  • SURVIVAL BUMP HUNTING METHOD FOR SUBGROUP IDENTIFICATION
    Wan Ke; Tanioka Kensuke; Minami Hiroyuki; Shimokawa Toshio; Mizuta Masahiro
    Bulletin of the Computational Statistics of Japan, 33, 1, 1, 28, Japanese Society of Computational Statistics, 2020, [Peer-reviewed]
    Japanese, Scientific journal, Efficacy and safety in randomized controlled clinical trials are often evaluated between active (treatment) and control groups. A limitation is that there may be disparity in (average) treatment effect between subjects. Statistical approaches to determine subgroups with higher treatment effect so to distinguish subgroups have therefore been developed.
    Threshold-based methods, such as CART (Classification and Regression Trees; Breiman et al., 1984), have been applied as methods of subgroup identification. Dusseldorp & Mechelen (2013) proposed the `QUINT' method, which uses Cohen's effect size as evaluation criteria. Meanwhile, tree-structured methods are a statistical approach, in which treatment effects are classified into similar subgroups, and are therefore not necessarily a suitable method for identifying responders.
    The present study instead focuses on the PRIM method (Patient Rule Induction Method; Friedman & Fisher, 1999). Previously reported PRIM-based methods (SPRIM; Kehl & Ulm, 2006) assumed proportional hazards between active (treatment) and control groups.
    We consider the application of a different PRIM method for survival data without such assumption of proportional hazards. It applies restricted mean survival time (RMST) to evaluate the measurement of treatment effect and is used to establish the subgroups to minimize or maximize any differences in treatment effect. The effectiveness of this method is illustrated through a practical example; a small-scale simulation suggests that it can be used to identify more appropriate subgroups than the SPRIM and SIDES (Lipkovich et al., 2011) methods.
  • AN APPLICATION OF SYMBOLIC DATA ANALYSIS FOR META-ANALYSIS
    Takagi Ryo; Minami Hiroyuki; Mizuta Masahiro
    Bulletin of the Computational Statistics of Japan, 32, 2, 95, 103, Japanese Society of Computational Statistics, 2020, [Peer-reviewed]
    Japanese, Scientific journal, The aim of this study is to propose a method to analyze objects in meta-analysis in symbolic data analysis, and represent an application for real data. Meta-analysis aims to obtain a higher level of evidence by integrating multiple studies. Then, it is important to take heterogeneity among the studies into consideration. When we regard the studies as concepts in symbolic data analysis, we can find out factors which may bring heterogeneity. The study proposes a method of hierarchical symbolic clustering when concepts are described as two-way contingency tables. The target data from multiple randomized trials on acupuncture clinical trials contain two-way contingency tables of outcomes and interventions, and individual patient data. The results reveal interpretable clusters and detect some variables for heterogeneity.
  • Exploratory Data Analysis and Cyber Security
    MINAMI Hiroyuki
    IEICE Technical Report, 119, 197, 17, Sep. 2019, [Invited]
    Japanese, Symposium
  • Analyses of Serum Fatty Acids and Vitamin D with Dimension Reduction Methods
    Y. Chen; R. Shrestha; Z. Chen; H. Chiba; S. Hui; E. Okada; S. Ukawa; T. Nakagawa; K. Nakamura; A. Tamakoshi; Y. Komiya; H. Minami; M. Mizuta
    Book of Abstracts, European Conference on Data Analysis (ECDA) 2019, 39, Mar. 2019, [Peer-reviewed]
    English, International conference proceedings
  • Statistical Analysis of Unauthorized Internet Access Log Data and its Interpretation
    MINAMI Hiroyuki
    Book of Abstracts, European Conference on Data Analysis (ECDA) 2019, 77, Mar. 2019, [Peer-reviewed]
    English, International conference proceedings
  • Empirical Study on Analysis of Unauthorized-Access Log Data and its Visual Output
    MINAMI Hiroyuki
    Book of Abstracts, European Conference on Data Analysis (ECDA) 2018, 86, Aug. 2018, [Peer-reviewed]
    English, International conference proceedings
  • Detecting Suicide Hotspots by Ages and Cities
    Q. Li; Y. Komiya; H. Minami; M. Mizuta
    Proceedings of Korea University and Hokkaido University 5th Workshop in Statistics, 35, 36, Feb. 2018
    English, International conference proceedings
  • An Application of Functional Data Analysis in Fatty Acids Concentration
    Y.Chen; Y. Komiya; H. Minami; M. Mizuta
    Proceedings of Korea University and Hokkaido University 5th Workshop in Statistics, 65, 68, Feb. 2018
    English, International conference proceedings
  • Symbolic Hierarchical Clustering for Two-way Contingency Tables
    R. Takagi; H. Minami; M. Mizuta
    Proceedings of Korea University and Hokkaido University 5th Workshop in Statistics, 17, 18, Feb. 2018
    English, International conference proceedings
  • Empirical Study on the Analysis of Brute-force Cyber Attack Logs
    H. Minami
    Proceedings of Korea University and Hokkaido University 5th Workshop in Statistics, 75, 76, Feb. 2018
    English, International conference proceedings
  • Symbolic data analytical approach to unauthorized-access logs
    Hiroyuki MINAMI; Masahiro MIZUTA
    Proc. of the 10th Conference of the IASC-ARS/68th Annual NZSA Conference, Dec. 2017, [Peer-reviewed], [Invited]
    English, International conference proceedings
  • Meta-analysis with symbolic data analysis and its application for clinical data
    R. Takagi; H. Minami; M. Mizuta
    Proceedings of the 2017 IASC-ARS/NZSA Conference (e-print), Dec. 2017, [Peer-reviewed]
    English, International conference proceedings
  • 学会設立30周年記念号の刊行にあたって(お詫びと御礼)
    中村 永友; 南 弘征
    計算機統計学, 30, 2, 103, 104, 日本計算機統計学会, 2017
    Japanese
  • Exploratory Methods for Joint Distribution Valued Data and Their Application
    Igarashi Kazuto; Minami Hiroyuki; Mizuta Masahiro
    COMMUNICATIONS FOR STATISTICAL APPLICATIONS AND METHODS, 22, 3, 265, 276, May 2015, [Peer-reviewed]
    English, Scientific journal
  • CLASSIFICATION OF CUSTOMERS AND ANALYSIS OF PURCHASE TENDENCY IN A GROUP-BUYING COUPON SITE
    Igarashi Kazuto; Matsui Yusuke; Minami Hiroyuki; Mizuta Masahiro
    Bulletin of the Computational Statistics of Japan, 28, 2, 139, 146, Japanese Society of Computational Statistics, 2015, [Peer-reviewed]
    Japanese, Scientific journal, In this paper, we propose a method for extracting useful information of the customer's purchase tendency from data recorded in a group-buying coupon site. To begin with, we classify the customers using cluster analysis according to categories of purchased coupon. We also show the relationship between the customer's attributes and the categories of coupons using correspondence analysis. Through comparison of the results for all customers and for the customers in each cluster, we discuss the in-depth interpretation.
  • Moving Functional k-means Clustering and Its Application
    Matsui Yusuke; Komiya Yuriko; Minami Hiroyuki; Mizuta Masahiro
    Bulletin of Data Analysis of Japanese Classification Society, 4, 1, 43, 55, Japanese Classification Society, 2015, [Peer-reviewed]
    Japanese, Scientific journal,

    A novel functional clustering method for domain-dependent functional data is proposed. The functional data are defined on sequential domains and they might have particular clusters on each domain. Our approach, Moving Functional k-means Clustering, is to classify the peculiar domains based on each sequential set of clusters. A typical functional clustering with fixed number of clusters is applied to the domains, then we relabel the set of clusters to keep the previous labels as many as possible. We demonstrate our method with large scale sensing data of environmental radio activity level in Fukushima Prefecture.

  • Symbolic Cluster Analysis for Distribution Valued Dissimilarity
    Yusuke Matsui; Hiroyuki Minami; Masahiro Misuta
    COMMUNICATIONS FOR STATISTICAL APPLICATIONS AND METHODS, 21, 3, 225, 234, May 2014
    English, Scientific journal
  • Comparison of two distribution valued dissimilarities and its application for symbolic clustering
    Yusuke Matsui; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Studies in Classification, Data Analysis, and Knowledge Organization, 46, 37, 46, Kluwer Academic Publishers, 2014, [Peer-reviewed]
    English, International conference proceedings
  • DETECTION FOR IRREGULAR DOMAIN OF MULTI DIMENSIONAL VALUED FUNCTIONAL DATA AND ITS APPLICATION TO SENSING DATA
    Matsui Yusuke; Komiya Yuriko; Minami Hiroyuki; Mizuta Masahiro
    Bulletin of the Computational Statistics of Japan, 27, 2, 65, 77, Japanese Society of Computational Statistics, 2014, [Peer-reviewed]
    Japanese, Scientific journal, In this paper, we propose a method for identifying an irregular domain of multi dimensional data that are observed continuously and its application to sensing data. We partition the data into three subdomains on each dimension - "irregularly high valued domains", "irregularly low valued domains" and "other domains". Then we detect the subdomains in which the irregularities are observed simultaneously. We demonstrate our method with an actual example using sensing data of environmental radio activity level in Fukushima Prefecture.
  • SYMBOLIC PRINCIPAL COMPONENT ANALYSIS FOR MODAL INTERVAL-VALUED DATA
    HAMADA ATSUSHI; MINAMI HIROYUKI; MIZUTA MASAHIRO
    計算機統計学, 26, 1, 3, 16, 日本計算機統計学会, 12 Dec. 2013, [Peer-reviewed]
    Japanese, Scientific journal, In this paper, we propose Principal Component Analysis (PCA) for modal interval-valued data. Professor Diday proposed Symbolic Data Analysis (SDA) which aggregate large datasets without loss of essential information. The aggregated data is called symbolic data. We focus on PCA in the SDA framework. Most studies on PCA in SDA assume interval-valued data. However, if we use interval-valued data as an aggregation, the variation of the original data is often ignored. The variation might have essential information and should be utilized. We introduce modal interval-valued data represented by histograms to PCA. We evaluate and compare the results of the proposed method and those of conventional methods through simulations. In addition, we apply our method to real data about TV drama and show its availability.
  • Non-hierarchical Symbolic Clustering for Distribution Valued Data and Its Application for Big Traffic Data
    YONEMORI CHIKARA; MINAMI HIROYUKI; MIZUTA MASAHIRO
    日本統計学会誌, 43, 1, 59, 67, 日本統計学会, Sep. 2013, [Peer-reviewed]
    Japanese, Scientific journal, We propose a novel non-hierarchical symbolic clustering for distribution valued data for an approach of big data analysis. The data are represented as probability distributions in symbolic data analysis (SDA). SDA is an effective approach that enables to analyze complex and various data that consist of high-level data expression including modal value, interval, distribution. In the proposed method, we implement clustering considering the characteristic in big data 'variety' by utilizing data representation in SDA. We conducted experiments for big traffic data to show the effectiveness of our method. In the experiments, we demonstrated clustering by road characteristic with distribution valued data made of GPS measurements by car. We also offered that the results give better suggestions on the traffic.
  • Extraction of Contact-Keywords History from GPS Data and its Application for the Inference of User Preferences
    FUJITA Masanori; TEZUKA Hirohisa; MUTO Shinyo; MINAMI Hiroyuki; MIZUTA Masahiro
    Kodo Keiryogaku, 40, 1, 3, 15, The Behaviormetric Society of Japan, 28 Mar. 2013, [Peer-reviewed]
    Japanese, Scientific journal, This paper proposes a method to make and analyze "contact-keywords history" from GPS logging data. We develop the method to infer gourmet preferences. The history consists of time-series attributes related to spots (shops, restaurants, etc.) in GPS logging data. These days, many contents-search or recommendation services for large volumes of contents in the Internet are available. Typically, the systems on these services need to have quite precise user requests before they can offer desirable contents. It is a hard work for a user to give enough information for the systems. Since most portable devices including mobile phones have GPS function, we can easily get GPS data and utilize them for user preferences. We evaluate the proposed method through practical GPS data.
  • ビッグデータ解析環境に関する考察(セッション8A 特別セッション「ビッグデータ」)
    南 弘征
    日本計算機統計学会シンポジウム論文集, 27, 249, 252, 日本計算機統計学会, 2013
    Japanese
  • 共同購入型クーポンサイトにおける顧客属性と購買傾向に関する考察(セッション3B スタディーグループセッション「データカフェ」)
    五十嵐 千人; 高倉 潤也; 鈴木 和之; 高畑 優修; 藤崎 稔晃; 松井 佑介; 南 弘征
    日本計算機統計学会シンポジウム論文集, 27, 93, 96, 日本計算機統計学会, 2013
    Japanese
  • テキストマイニングによるネット販売サイトのキャッチフレーズ解析(セッション3B スタディーグループセッション「データカフェ」)
    高倉 潤也; 五十嵐 千人; 鈴木 和之; 高畑 優修; 藤崎 稔晃; 松井 佑介; 南 弘征
    日本計算機統計学会シンポジウム論文集, 27, 97, 98, 日本計算機統計学会, 2013
    Japanese
  • SDA framework is the tool for Big Data Analysis?
    MINAMI Hiroyuki; Masahiro MIZUTA
    Book of Abstracts. 3rd Workshop in Symbolic Data Analysis, 21, Nov. 2012, [Peer-reviewed]
    English, International conference proceedings
  • Spot Recommendation Through Integrated Analysis of GPS Log and Browsing Log
    藤田将成; 伊藤浩二; 小林稔; 南弘征; 水田正弘
    データ分析の理論と応用, 2, 1, 53, 68, 日本分類学会, 01 Sep. 2012, [Peer-reviewed]
    Japanese, Scientific journal
  • Analysis of distribution valued dissimilarity data
    Masahiro Mizuta; Hiroyuki Minami
    Studies in Classification, Data Analysis, and Knowledge Organization, 23, 28, Springer, 2012, [Peer-reviewed]
    English, International conference proceedings
  • Toward complementary application with Symbolic Data Analysis and Rough Set theory
    MINAMI Hiroyuki
    Joint Meeting of the 2011 Taipei International Statistical Symposium and 7th Conference of the Asian Regional Section of the IASC, 17, b4, 1, Dec. 2011, [Invited]
    English, International conference proceedings
  • A Clustering Method for Distribution Valued Dissimilarities
    Yusuke Matsui; Masahiro Mizuta; Hiroyuki Minami; Yuriko Komiya
    Proceedings of Joint Meeting of the Korea-Japan Conference of Computational Statistics and the 25th Symposium of Japanese Society of Computational Statistics, 101, 102, Nov. 2011
    English
  • A statistical survey on bulk emails with symbolic data analysis
    Hiroyuki MINAMI
    Collection of Abstracts, IFCS Symposium and GfKl/DAGM Conference Talks, 140, Aug. 2011, [Peer-reviewed]
    English
  • A DERIVATIVE OF INTERVAL-VALUED FUNCTIONAL DATA AND ITS APPLICATION
    Ikeda Tomoyasu; Komiya Yuriko; Minami Hiroyuki; Mizuta Masahiro
    Bulletin of the Computational Statistics of Japan, 23, 2, 113, 123, Japanese Society of Computational Statistics, Apr. 2011, [Peer-reviewed]
    Japanese, Scientific journal, In this paper, we propose a derivative of interval-valued functional data, and apply it to cluster analysis. Symbolic Data Analysis (SDA) proposed by Diday is a new approach for analyzing various data which are represented by symbolic concept including numerical data, interval-valued data and modal-valued data. Functional Data Analysis (FDA) developed by Ramsay is an approach for analyzing functional data. In FDA, we can make good use of "smoothness" in datasets. Toyoda et al. (2009) proposed a clustering method for interval-valued functional data which is an extension of functional data. W...
  • COMMON PRINCIPAL COMPONENTS MODEL FOR SYMBOLIC DATA
    KATAYAMA Kotoe; MINAMI Hiroyuki; MIZUTA Masahiro
    Journal of the Japanese Society of Computational Statistics, 23, 1, 41, 50, Japanese Society of Computational Statistics, 01 Dec. 2010, [Peer-reviewed]
    English, Scientific journal, This paper proposes a dimension reduction technique in the framework of symbolic data analysis (SDA). Recent advances in technology have increased the complexity of datasets, and today, their size is much larger than it was in the past decade. Most statistical methods do not have sufficient power to analyze these datasets. SDA was proposed by Diday at the end of the 1980s and is a new approach for analyzing huge and complex data.
    SDA examines “symbolic data”, which consist of concepts. A concept consists of not only values but also “higher-level units” such as an interval and a distribution. Their combination can also be represented as a kind of a concept. This implies that complex data can be formally handled in the framework of SDA. However, there are very few studies based on this simple idea. Therefore, practical methods should be developed to apply this idea to solve problems in the real world. In this study, we focus on the case in which a concept contains some subsets (the concept acts as a typical complex dataset) and develop a new method to analyze this dataset directly using SDA.
    In this paper, we propose a dimension reduction technique in the framework of SDA, especially for a group structure, and introduce a numerical example.
  • Semi-supervised Discriminant Analysis for Interval-valued Data
    Kenji TOYODA; Hiroyuki MINAMI; Masahiro MIZUTA
    19th International Conference on Computational Statistics, 176, Aug. 2010, [Peer-reviewed]
    English
  • Empirical Studies on the Analysis of a Vast Amount of the Internet Traffic Data in Japan
    Hiroyuki MINAMI
    Abstracts of the 3rd German-Japanese Workshop, University of Karlsruhe, 22, Jul. 2010, [Peer-reviewed]
    English
  • A Knowledge Engineering Applied Symbolic Data Analysis
    Hiroyuki MINAMI; Masahiro MIZUTA
    Abstracts of the 34th Annual Conference of the German Classification Society (Gfkl), 127, Jul. 2010, [Peer-reviewed]
    English
  • Case Study on Trends of Consumers from House Card Data
    Naoki KISHIMOTO; Yuriko KOMIYA; Hiroyuki MINAMI; Masahiro MIZUTA
    Proceedings for the Spring Conference, 2010, The Korean Statistical Society, 46, May 2010, [Peer-reviewed]
    English
  • An Extension of Correspondence Analysis for Functional Data
    Kazuya KONDO; Kenji TOYODA; Hiroyuki MINAMI; Masahiro MIZUTA
    Proceedings for the Spring Conference, 2010, The Korean Statistical Society, 37, May 2010, [Peer-reviewed]
    English
  • HIERARCHICAL SYMBOLIC CLUSTERING FOR DISTRIBUTION VALUED DATA
    Katayama Kotoe; Minami Hiroyuki; Mizuta Masahiro
    Bulletin of the Computational Statistics of Japan, 22, 2, 83, 89, Japanese Society of Computational Statistics, May 2010, [Peer-reviewed]
    Japanese, Scientific journal, We propose a hierarchical clustering in the framework of Symbolic Data Analysis (SDA). SDA was proposed by Diday at the end of the 1980s and is a new approach for analysing huge and complex data. In SDA, an observation is described by not only numerical values but also "higher-level units"; sets, intervals, distributions, etc. Most SDA works have dealt with only intervals as the descriptions. In this paper, we propose a hierarchical clustering for distribution valued data and show its effectiveness through a numerical simulation.
  • An Extension of Functional PCA to Interval-Valued Functional Data
    IKEDA Tomoyasu; KOMIYA Yuriko; MINAMI Hiroyuki; MIZUTA Masahiro
    Japanese Journal of Applied Statistics, 39, 1, 21, 33, Japanese Society of Applied Statistics, Apr. 2010, [Peer-reviewed]
    Japanese, Scientific journal, We discuss an extension of Functional Principal Component Analysis (Functional PCA) to Symbolic Data Analysis (SDA).<BR>SDA proposed by Diday is a new approach for analyzing datasets which are too large and complex to handle with conventional methods. In SDA, an observation is represented by <I>symbolic concept</I> including numerical, interval-valued and modal-valued data. Symbolic PCA methods have been studied as dimension reduction techniques, which are mainly applied to interval-valued data.<BR>Another approach for a huge variety of datasets is Functional Data Analysis (FDA), developed ...
  • Visualization of influence patterns in Linear Subspace Method
    Kuniyoshi Hayashi; Hiroyuki MINAMI; Masahiro MIZUTA
    Joint Meeting of Japan-Korea Special Conference of Statistics and the 2nd Japan-Korea Statistics Conference of Young Researchers, 81, 88, Feb. 2010, [Peer-reviewed]
    English
  • On multiple-case diagnostics in linear subspace method
    Kuniyoshi Hayashi; Hiroyuki Minami; Masahiro Mizuta
    Proceedings of COMPSTAT 2010 - 19th International Conference on Computational Statistics, Keynote, Invited and Contributed Papers, 493, 500, 2010, [Peer-reviewed]
    English, International conference proceedings
  • Detection of Influential Low Usability Web Pages Using Variation of Maximum Log-Likelihood
    Kuniyoshi Hayashi; Toshiya Yamada; Noboru Nakamichi; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Proc. The 3rd International Symposium on Global COE Program of Center for Next-Generation Information Technology Based on Knowledge Discovery and Knowledge Federation (GCOE-NGIT 2010), 159, 160, Jan. 2010
    English, International conference proceedings, Sapporo, Japan [Poster]
  • Webユーザビリティ定量評価における局所影響診断の適用について -(分析結果に対する観測値影響パターンのの知識発見)-
    林邦好; 高橋一磨; 中道上; 小宮由里子; 南弘征; 水田正弘
    第2回若手研究者支援のための産学協同GCOE国内シンポジウムダイジェスト集, 49, 50, 30 Sep. 2009
    Japanese, [Poster]
  • AN EXTENSION OF RELATIVE PROJECTION PURSUIT TO FUNCTIONAL DATA
    Hiro Shintaro; Minami Hiroyuki; Mizuta Masahiro
    Bulletin of the Computational Statistics of Japan, 21, 1, 15, 28, Japanese Society of Computational Statistics, May 2009, [Peer-reviewed]
    Japanese, Scientific journal, In this paper, we propose a functional relative projection pursuit as an extended method of relative projection pursuit (Mizuta, 2002a) for functional data analysis. This method finds an 'interesting' structure of subset data in a low dimensional projection space of the functional data compared with the structure of the superset data. For example, it can detect some different part of functions such as a time cycle and amplitude between the subset functional data and the superset functional data. Projection pursuit (Friedman & Tukey, 1974) is one of the dimension reduction methods to search ...
  • 音楽著作権使用料分配のための楽曲利用状況推定法について
    加藤順; 原田高; 南弘征; 水田正弘
    電子情報通信学会論文誌 D, J92-D, 4, 455, 466, The Institute of Electronics, Information and Communication Engineers, Apr. 2009, [Peer-reviewed]
    Japanese, Scientific journal, 放送番組で利用される音楽著作権処理のために,音楽の特徴(フィンガープリント)を利用した楽曲特定サービスの検討が進んでいる.音楽使用料の権利者への分配を正確に行うためには,できるだけ多くの放送番組での楽曲利用を把握する必要があるが,全放送番組を網羅することはコスト面から見て現実的ではない.本論文では,効率的に楽曲利用状況を推定するためのサンプリング手法を提案し,サンプリングモデルの検討及び実データを用いたシミュレーションにより,その有効性を確認した.更に,分配誤差の指標を提案し,その指標を用いて,コストに直接的に関係するサンプリング割合と,分配誤差の関係を明らかにした.例えば,サンプリング割合が1/2以上の場合,楽曲の利用割合で換算して7割以上の楽曲に対する分配誤差を10%以下に抑えられることなどの傾向を確認した.これらの結果は,サンプリングに基づく楽曲特定サービスの導入のための判断基準を与える.
  • Comparative studies on real applications for Internet Control Management Protocol data with Functional and Symbolic Data Analysis.
    Hiroyuki MINAMI
    Proceedings of IFCS2009 / GFKL, 115, Mar. 2009, [Peer-reviewed]
    English
  • Diagnostics of Linear Discriminant Analysis for Detecting Low Usability Web Pages
    Kuniyoshi Hayashi; Toshiya Yamada; Noboru Nakamichi; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Proc. The 2nd International Symposium on Global COE Program of Center for Next-Generation Information Technology Based on Knowledge Discovery and Knowledge Federation (GCOE-NGIT 2009), 169, 170, 20 Jan. 2009
    English, International conference proceedings, Sapporo, Japan [Poster]
  • Principal Component Analysis for Modal Interval-Valued Data
    Atsushi Hamada; Hiroyuki Minami; Masahiro Mizuta
    Proceedings of IASC2008 (CD-ROM), 264, Dec. 2008, [Peer-reviewed]
    English
  • Auto Regression Analysis for Symbolic Distribution-Valued Data
    Kazuma Takahashi; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Proceedings of IASC2008 (CD-ROM), 265, Dec. 2008, [Peer-reviewed]
    English
  • An analysis of Layer 2 network monitoring data with huge-data oriented statistical techniques
    Hiroyuki Minami
    Proceedings of IASC2008 (CD-ROM), 258, Dec. 2008, [Peer-reviewed]
    English
  • An Extension of Functional PCA to Interval-Valued Functional Data
    Tomoyasu Ikeda; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Proceedings of IASC2008 (CD-ROM), 263, Dec. 2008, [Peer-reviewed]
    English
  • インタラクション定量分析における観測値の影響診断 -(Webユーザビリティ評価を中心として)-
    林邦好; 山田俊哉; 中道上; 小宮由里子; 南弘征; 水田正弘
    第1回若手研究者支援のための産学協同GCOE国内シンポジウムダイジェスト集, 27, 28, Oct. 2008
    Japanese
  • Symbolic Multidimensional Scaling and Its Application to Internet Traffic Data
    Hiroyuki Minami; Masahiro Mizuta
    Book of Abstract, CompStat2008, 171, Aug. 2008, [Peer-reviewed]
    English
  • Toward a Complementary Application between Symbolic Data Analysis and Knowledge Engineerning
    Hiroyuki Minami
    Proceedings of the 9th Japan-China Symposium on Statistics, 187, 190, Sep. 2007
    English
  • Comparative study on a criterion to determine the number of classes in Latent Class Analysis
    Kotoe Katayama; Hiroyuki Minami; Masahiro Mizuta
    Bulletin of the International Statistical Institute, 56th Session(CD-ROM), 1, 4, Aug. 2007, [Peer-reviewed]
    English
  • A criterion to determine the number of classes in Latent Class Analysis
    Kotoe Katayama; Hiroyuki Minami; Masahiro Mizuta
    The 72nd Annual Meeting of the Psychometric Society. IMPS2007, 41335, Jul. 2007, [Peer-reviewed], [Invited]
    English
  • Linearly Restricted Principal Components in k Groups
    K. Katayama; A. Suzukawa; H. Minami; M. Mizuta
    KNEMO’06, Knowledge Extraction and Modeling (CD-ROM), (CP4-1), Sep. 2006, [Peer-reviewed]
    English
  • Regularization Parameter Selection in the Group Lasso
    T. Shimamura; H. Minami; M. Mizuta
    KNEMO’06, Knowledge Extraction and Modeling (CD-ROM), (CP5-2), Sep. 2006, [Peer-reviewed]
    English
  • An application of relative projection pursuit for functional data to human growth
    Shintaro Hiro; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Proceedings in Computational Statistics 2006, 1113, 1120, Aug. 2006, [Peer-reviewed]
    English, Scientific journal
  • Generalized Linear Model with the Lasso and Bayesian information - criteria for selection of the regularization parameter
    Teppei Shimamura; Hiroyuki Minami; Masahiro Mizuta
    17th Symposium of IASC on Computational Statistics, Book of Abstracts, 287, Aug. 2006, [Peer-reviewed]
    English
  • A generalization of common principal components analysis
    Kotoe Katayama; Akio Suzukawa; Hiroyuki Minami; Masahiro Mizuta
    17th Symposium of IASC on Computational Statistics, Book of Abstracts, 154, Aug. 2006, [Peer-reviewed]
    English, Scientific journal
  • Empirical Study on the ICMP traffic data via the Internet
    Hiroyuki Minami; Masahiro Mizuta
    IFCS2006 Abstracts, 116, Jul. 2006
    English
  • Statistical Approach on Internet Traffic Data
    Hiroyuki MINAMI; Masahiro MIZUTA
    The 5th IASC Asian Conference on Statistical Computing(IascAsian05), Dec. 2005, [Peer-reviewed]
    English
  • Analysis of Visitor Trends with Web Logs
    Ko Harada; Hiroyuki Minami; Masahiro Mizuta
    Bulletin of the International Statistical Institute, 55th Session, 274, Apr. 2005, [Peer-reviewed]
    English
  • SPAM Filtering based on E-mail Headers and the Comparison with Bayesian Filtering on its Contents
    Hiroyuki Minami
    Bulletin of the International Statistical Institute, 55th Session, 276, Apr. 2005, [Peer-reviewed]
    English
  • Multidimensional Relative Projection Pursuit
    HIRO Shintaro; KOMIYA Yuriko; MINAMI Hiroyuki; MIZUTA Masahiro
    Japanese journal of applied statistics., 33, 3, 225, 241, 応用統計学会, 25 Mar. 2005, [Peer-reviewed]
    Japanese, Scientific journal
  • Nonlinear Regression Modeling Strategy Based on Radial Basis Function Networks and Absolute Shrinkage
    Teppei Shimamura; Hiroyuki Minami; Masahiro Mizuta
    Proceedings of IFCS2004, Aug. 2004, [Peer-reviewed]
  • Parallel data analysis and its implementation on PC cluster
    Hiroyuki Minami; Masahiro Mizuta
    8TH WORLD MULTI-CONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL V, PROCEEDINGS, V, 75, 79, 2004, [Peer-reviewed], [Lead author]
    English, International conference proceedings
  • Evaluation of Execution Time on Data Analysis with Parallel Virtual Machine
    Hiroyuki MINAMI; Yuriko KOMIYA; Masahiro MIZUTA
    Journal of the Japanese Society of Computational Statistics, 15, 2, 193, 199, Dec. 2003, [Peer-reviewed]
  • Adaptive Spam Filtering with Text Mining
    Yusuke Sato; Hiroyuki Minami; Masahiro Mizuta
    Bulletin of the International Statistical Institute, 54th Session, 2, 371, 372, Aug. 2003, [Peer-reviewed]
  • Penalized Regression Modeling for Gene Expression Data Analysis
    Teppei Shimamura; Hiroyuki Minami; Yuriko Komiya; Mitsuhiro Tada; T. Moriuchi; Masahiro Mizuta
    Bulletin of the International Statistical Institute, 54th Session, 2, 426, 427, Aug. 2003, [Peer-reviewed]
  • Hall Type Projection Index for Relative Projection Pursuit
    Shintaro Hiro; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Bulletin of the International Statistical Institute, 54th Session, 1, 492, 493, Aug. 2003, [Peer-reviewed]
  • Empirical Study of Parallel-oriented Statistical Procedure
    Hiroyuki Minami
    Bulletin of the International Statistical Institute, 54th Session, 2, 51, 52, Aug. 2003, [Peer-reviewed]
  • Projection Index with Empirical Distribution Function
    HIRO Shintaro; KOMIYA Yuriko; MINAMI Hiroyuki; MIZUTA Masahiro
    Japanese journal of applied statistics, 32, 1, 17, 29, Japanese Society of Applied Statistics, 31 Jul. 2003, [Peer-reviewed]
    Japanese, In this paper, we propose a new method of projection pursuit which extends a concept of 'interestingness' and develop a projection index for it.<BR>It is generally difficult to see the structure of data when the dimension of data is high. Therefore, many studies have been reported on the methods which reduce high dimensional data into a smaller number of dimensions. Among those methods, projection pursuit was developed by Friedman and Tukey (1974) in order to search for an 'interesting' linear projection of multidimensional data. The feature 'interestingness' is defined as the difference between a distribution of target data and normal distribution. To measure this feature, many projection indices have been proposed to formulate the 'interestingness' mathematically.<BR>However, measuring the difference from normal distribution by such projection indices does not always give 'interestingness' because 'interesting' structure depends on data or on purposes of the analysis.<BR>Thus, we develop a method with a new projection index that a user can pre-define an 'interesting' structure according to the feature of the data or the purpose of the analysis. Our index measures a distance between the distribution of the projection of target data and that of the projection of pre-defined 'uninteresting' data. We show the effectiveness of the proposed method with numerical examples and actual data.
  • Discrimination with Variable Selection for the Analysis of Microarray Data
    Teppei Shimamura; Hiroyuki Minami; Yuriko Komiya; Mitsuhiro Tada; T. Moriuchi; Masahiro Mizuta
    Proceedings of the 4th Conference of the Asian Regional Section of the International Association for Statistical Computing, 118, 121, Dec. 2002, [Peer-reviewed]
  • Comparative Study of Performances of PC Cluster Environments in Statistical Computing
    Hiroyuki MINAMI; Masahiro MIZUTA
    Proceedings of the 4th Conference of the Asian Regional Section of the International Association for Statistical Computing, 151, 154, Dec. 2002, [Peer-reviewed]
  • Analysis of Spam Mail Based on Email Contents
    Yusuke Sato; Hiroyuki Minami; Masahiro Mizuta
    Proceedings of the 4th Conference of the Asian Regional Section of the International Association for Statistical Computing, 12, 14, Dec. 2002, [Peer-reviewed]
  • Search for the Effective Dimension Reduction Directions in Sliced Inverse Regression Model with ICA
    Atsushi Ogiya; Yuriko Komiya; Hiroyuki Minami; Masahiro Mizuta
    Proceedings of the 4th Conference of the Asian Regional Section of the International Association for Statistical Computing, 5, 8, Dec. 2002, [Peer-reviewed]
  • TOPICS RELATED TO HIGH PERFORMANCE COMPUTING ENVIRONMENT AND COMPUTATIONAL STATISTICS
    MINAMI HIROYUKI
    計算機統計学, 15, 2, 273, 281, 2002, [Peer-reviewed]
    Japanese, Scientific journal
  • 射影追跡層別逆回帰法における並列計算について
    松田光弘; 南弘征; 水田正弘
    応用統計学, 30, 2, 81, 90, Japanese Society of Applied Statistics, Nov. 2001, [Peer-reviewed]
    Japanese, Scientific journal, In the paper, we propose an SIRPP implementation on parallel processing with PVM, Parallel Virtual Machine.
    In regression analysis, a lot of explanatory variables sometimes make data analysis harder. When some explanatory variables are useless for predicting the value of response variable, the explanatory variable selection is very useful. But, if all of the explanatory variables are related to the response variable, we will search linear combinations of explanatory variables with projection pursuit regression and ACE (Alternating Conditional Expectations). These two methods are assumed special models.
    Sliced Inverse Regression (Li, 1991) is one of the approaches to reduce the number of explanatory variables in regression analysis. SIR does not get rid of some explanatory variables themselves but may reduce the dimension of a space of explanatory variables. It is based on the model (SIR model)
    y=f(β1x, β2x, …, βkx, ε),
    where x is the vector of p explanatory variables, βk(k=1, 2, …, K) are unknown row vectors, ε is independent of x, and f is an arbitrary unknown function on R[K+1].
    Mizuta (1999) proposed an algorithm for SIR model with projection pursuit, named SIRPP (Sliced Inverse Regression with Projection Pursuit). SIRPP has excellent performance in finding {βk}. However, the algorithm requires more computing power. It is a part of projection pursuit that the algorithm takes much time. In order to overcome the defect, we use PVM for SIRPP. We offer its effectiveness through numerical examples.
  • Evaluation of Execution Time on Data analysis with Parallel Virtual Machine
    Hiroyuki MINAMI; Yuriko KOMIYA; Masahiro MIZUTA
    Proceedings of the International Conference on New Trends in Computational Statistics with Biomedical Applications, 321, 326, Sep. 2001, [Peer-reviewed]
  • A Graphical User Interface for Parallel Data Analysis
    Atsushi Ogiya; Hiroyuki Minami; Masahiro Mizuta
    Bulletin of the International Statistical Institute, 53rd Session, 3, 191, 192, Aug. 2001, [Peer-reviewed]
  • Optimization Algorithm for Data Analysis with PVM
    Yuriko KOMIYA; Hiroyuki MINAMI; Masahiro MIZUTA
    Bulletin of the International Statistical Institute, 53rd Session, 3, 187, 188, Aug. 2001, [Peer-reviewed]
  • A Model of Execution time for the Bootstrap with PVM
    Hiroyuki MINAMI; Masahiro MIZUTA
    Bulletin of the International Statistical Institute, 53rd Session, 1, 241, 242, Aug. 2001, [Peer-reviewed]
  • A study on execution time for data analysis with parallel virtual machine.
    MINAMI HIROYUKI; MIZUTA MASAHIRO
    計算機統計学, 14, 2, 111, 122, Japanese Society of Computational Statistics, 2001, [Peer-reviewed], [Lead author]
    Japanese, Scientific journal, In this paper, we discuss a parallel execution about data analysis. Especially, we focus on the execution in parallel virtual computer consisted of many micro computers connected each other via network, because we can get this environment more easily than we get a super-computer. We have some simulations with the Bootstrap on PVM (Parallel Virtual Machine) environment, which is one of the popular software libraries to have some computers the virtual one, in order to investigate characteristics on execution time mainly. The parameters in the simulations are the number of Boostrap samples, the number of processors, calculation load in slave (expressed in a waiting time). The results show some important features and the relation between parameters and a round trip time, and we get a simple model with regression. This model is effective to consider execution time from an analysis on a virtual machine or how many personal computers we should prepare to analyze data in time, etc.
  • Empirical Study of Parallel Data Analysis
    Hiroyuki MINAMI; Masahiro MIZUTA
    Proceedings of the 10th Japan and Korea Joint Conference of Statistics, 131, 136, Dec. 2000, [Lead author]
    English
  • Multidimensional Scaling for Time-Dependent Data
    Masahiro MIZUTA; Hiroyuki MINAMI
    Proceedings of 7th International Conference on Neural Information Processing, 1, 646, 650, Oct. 2000, [Peer-reviewed]
  • A Parallel Implementation for the Bootstrap
    Hiroyuki MINAMI; Yuriko KOMIYA; Masahiro MIZUTA
    Proceedings of 7th International Conference on Neural Information Processing, 2, 1078, 1082, Oct. 2000
  • Visual Statistical System with Java
    Hiroyuki MINAMI; Masahiro MIZUTA
    Proceedings in Computational Statistics 2000, 69, 70, Aug. 2000, [Peer-reviewed]
  • An Algorithm with Projection Pursuit for Sliced Inverse Regression Model
    Masahiro MIZUTA; Hiroyuki MINAMI
    Proceedings of IFCS2000 as Data Analysis, Classification, and Related Methods, (H.A.L.Kiers, J.-P.Rasson, P.J.F.Groenen and M.Schader Eds.), Springer, 255, 260, Apr. 2000, [Peer-reviewed]
    English, International conference proceedings
  • インターネットを介した統計パネル討論の試行.(共著)
    計算機統計学, 12, 2, 137, 143, 2000, [Peer-reviewed]
    Japanese, Scientific journal
  • 統計教育はこれでいいのか!? : 新しい教育環境における計算機統計
    垂水 共之; 馬場 康維; 水田 正弘; 南 弘征; 森 裕一; 山本 義郎
    日本計算機統計学会シンポジウム論文集, 13, 0, 70, 71, 日本計算機統計学会, 1999
    Japanese
  • 動的な対散布図のJavaによる実装について
    山本義郎; 南弘征; 水田正弘; 佐藤義治
    計算機統計学, 11, 2, 113, 119, Japanese Society of Computational Statistics, 1998, [Peer-reviewed]
    Japanese, Scientific journal, 多次元データ解析において, 動的なグラフィックスの活用は極めて有効である.また, 昨今のインターネットの普及と相まって, 従来からの問題点である機種依存性の排除を可能としたJava言語が出現してきている. 本稿ではJavaについて概説するとともに, Javaで作成され, World Wide Web上で動作する「動的な対散布図」について, 報告する.
  • 仮説推論機構を用いた多変量データ解析支援システム
    南 弘征; 水田 正弘; 佐藤 義治
    応用統計学, 23, 2, 63, 79, 1994, [Peer-reviewed], [Lead author, Corresponding author]
    Japanese, Scientific journal
  • A framework of knowledge base for data analysis supporting system and its implementation.
    南弘征; 水田正弘; 佐藤義治
    計算機統計学, 6, 1/2, 37, 48, Japanese Society of Computational Statistics, Dec. 1993, [Peer-reviewed], [Lead author, Corresponding author]
    Japanese, Scientific journal, In this paper, we propose a framework based on "Frame Theory" for data analysis supporting system to construct the set of the knowledge for data analyses. There is some research to which the technique of knowledge processing to data analysis is applied. In the field of knowledge processing, a knowledge base and an inference function are to be independent for flexibility of a system. Most reports about the application, however, may not tell us that both parts are separated clearly. In this paper, we compare some methods about knowledge representation shown in the previous works for data analysis supporting systems and show that "Frame Theory" is the most available when we focus on the supporting function to select appropriate statistical procedures. We also offer how to implement the knowledge of data analysis on the framework and show two examples about supporting functions based on our knowledge base and the independence of the knowledge base and the inference functions through these examples.
  • A Knowledge supporting system for data analysis.(共著)
    Minami Hiroyuki; Mizuta Masahiro; Sato Yoshiharu
    Journal of the Japanese Society of Computational Statistics, 6, 1, 85, 97, Japanese Society of Computational Statistics, 1993, [Peer-reviewed], [Lead author, Corresponding author]
    English, Scientific journal, Most data analysis systems seem to be convenient for statisticians, but novices in statistics cannot master them easily since it is hard to select appropriate procedures for an observation. The application with the technique of knowledge processing to data analysis is an attractive idea, but there is little research on selecting procedures because of some difficulties in the construction of its knowledge. We shall discuss its difficulties and how to construct the knowledge of data analysis, and offer an implementation of the system with knowledge supports.
  • A refined algorithm of principal curves and the evaluation of its complexity.
    山下信之; 南弘征; 水田正弘; 佐藤義治
    計算機統計学, 5, 1, 33, 43, Japanese Society of Computational Statistics, May 1992, [Peer-reviewed]
    Japanese, Scientific journal, Relatively few methods have been proposed, that could be used in the fitting of curves to data without external criterion, i.e, data with no clear distinction between objective and explanatory variables. Hastie and Stuetzle (1989) have proposed Principal Curves as an algorithm to fit curves to such data. The algorithm repeats alternately a Projection Step and an Expectation Step until a convergence condition is satisfied. But in the Projection Step, we must search the nearest neighbouring points on the curve from each of N data points, thus computational complexity of order N^2 is necessary in the straightforward search. In this paper, we propose an effective algorithm for the Projection Step, which untilizes a recursive binary-tree search. We evaluate the computational complexities of the straiforward and the refined algorithm for some special cases, and show efficiency of our algorithm on some examples.
  • A graphical representation for MDS.
    水田正弘; 南弘征
    計算機統計学, 4, 2, 25, 30, Japanese Society of Computational Statistics, Dec. 1991, [Peer-reviewed]
    Japanese, Scientific journal, A new graphical method for multidimensional scaling is proposed in this paper. One of the most important standpoints in exploratory data analysis is graphical representation of residuals. For example, we can investigate a result of regression analysis with a scatter plot and vertical segments from data points to the regression line. In the case of multidimensional scaling, the relation of input data and configuration is usually verified by a scatter plot of input dissimilarities and distances among the configuration. However, the scatter plot can not show the objects that are not adequate. So, we propose a graphical method for multidimensional scaling, which represents the configuration and residuals at the same time. We can make use of the method to represent asymmetric dissimilarity data graphically. The source program of the proposed method is shown with S language. The source programs of Basic version and C version are also provided. The proposed method is illustrated with some data.
■ Other Activities and Achievements
■ Books and other publications
  • 統計科学百科事典
    日本統計学
    丸善出版, Dec. 2018, 9784621303108, Japanese, Dictionary or encycropedia, [Contributor]
  • 統計応用の百科事典
    松原望; 美添泰人, Webデータ解析
    丸善出版, Oct. 2011, 9784621083970, Japanese, Dictionary or encycropedia, [Contributor]
  • 統計データ科学事典
    杉山高一; 藤越康祝; 杉浦成昭; 国友直人, グラフィカルユーザインタフェース
    朝倉書店, Jun. 2007, 9784254121650, Japanese, Dictionary or encycropedia, [Contributor]
  • 確率モデルによるWebデータ解析法 - データマイニング技法からe-コマースまで
    水田 正弘; 南 弘征; 小宮 由里子
    水田正弘; 南 弘征; 小宮由里子, 森北出版, May 2007, 4627095317, 320, [Joint translation]
  • S‐PLUSによるデータマイニング入門
    水田 正弘; 南 弘征; 山本 義郎; 田澤 司
    森北出版, Jan. 2005, 4627845219, 192, [Joint work]
  • インターネットで広がる北海道のマルチメディア
    北海道新聞社(編), 1996, [Contributor]
  • fjの歩き方-インターネットニュースグループの世界-
    オーム社出版局, 1995, [Joint work]
■ Lectures, oral presentations, etc.
■ Syllabus
  • 先端データ科学特論, 2024年, 修士課程, 情報科学院
  • 先端データ科学特論, 2024年, 博士後期課程, 情報科学研究科
  • 先端データ科学特論, 2024年, 博士後期課程, 情報科学院
  • 確率過程とデータ解析, 2024年, 学士課程, 工学部
  • 情報セキュリティ, 2024年, 学士課程, 工学部
  • 専門横断科目（一般科目）：複合領域, 2024年, 学士課程, 専門横断科目
■ Affiliated academic society
■ Research Themes
  • 集約的データ解析法による不正アクセス履歴の分析とサイバー攻撃予測への応用
    科学研究費助成事業 基盤研究(B)
    Apr. 2023 - Mar. 2028
    南 弘征; 棟朝 雅晴; 馬場 健一; 水田 正弘; 山岡 克式
    日本学術振興会, 基盤研究(B), 北海道大学, Principal investigator, 23H03392
  • A Study of Exploratory Meta-Analysis Based on Analytic Meta-Analysis and SDA
    Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
    01 Apr. 2018 - 31 Mar. 2023
    水田 正弘; 清水 信夫; 南 弘征; 松井 佑介; 飯島 弘章
    本年度は、以下の事項について研究を推進した。
    （１）メタ解析において仮定するモデルを選択するための異質性の尺度として、QおよびI-squareがよく知られている。これらの値によって母数モデルまたは変量モデルの妥当性を評価することができる。しかし、研究成果ごとの「異質性」には様々な可能性がある。系統的な非線形関係が存在する場合と外れ値が存在する場合では、対処法が異なる。それらの異質性を同一に扱うのは不適切である。そこで、本年度は、異質性を単一の数値で表現するのではなく、シンボリック分布値データとして解析し、各研究成果を散布図などで表現する方法を提案した。
    （２）効果変数が複数設定されている場合におけるメタ解析として、多変量メタ解析(Multivariate Meta-Analysis)が提唱されている。例えば、効果変数として「生存率」と「無病生存率」の2つを設定した場合、これらの値は独立ではない。そこで、それらの相関係数を考慮した手法が検討されている。これをシンボリックデータ解析の立場から解釈すると、構造を有する分布値データと扱うことができる可能性がある。そこで、新たな理論構築を目指した。
    （３）典型的なメタ解析では、介入(Interventions、薬や治療法など）の有無またはプラセボや既存の治療法との比較が中心課題となっている。しかし、現実には、複数の介入方法が存在することが多い。そこで、複数の不完全な比較研究を用いて、全体の評価を実施するネットワークメタ解析(Network Meta-Analysis)が検討されている。そこでは、介入による効果の順序付けなどが問題となる。ネットワークメタ解析における発見的な手法、すなわち探索的ネットワークメタ解析を構築するために、シンボリックデータ解析のアプローチを適用する方法を検討した。
    Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), Hokkaido University, 18H03207
  • 情報ネットワーク研究用標準トラヒックデータ集策定に向けた企画調査
    科学研究費補助金(基盤研究(C))
    2006 - 2006
    山岡 克式; 南 弘征; 馬場 健一; 山本 義郎
    研究用標準トラヒックデータ集策定を目的とした研究体制構築を目指して,本企画調査では(1)標準化項目の調査,(2)標準データ策定手順の調査,(3)標準化団体の調査,の3項目を実施した.(1)標準化項目の調査学会その他研究者の集まる場において,研究者のニーズを調査し,標準データ集に要求されるデータ種類やパラメータ項目の把握を行った.(2)標準データ策定手順の調査標準画像その他の研究用標準データ策定を実際に行った団体や作業メンバーに対して調査を行い,各標準データ策定作業の手順,作業内容,作業量,作業体制,および問題点を把握した.(3)標準化作業の調査国際標準化団体に対して調査を行い,標準化手続きや作業内容を把握した.以上の調査の結果,標準データは,何らかの標準化団体において,その標準化を目的とした特定の目的(特定の効果の測定や評価)のために作成されるケースが大半であり,その目的を達成するためにある意味恣意的なデータを新規で作成して標準とする場合と,既存データの中からその目的に合致する適切なデータを標準として選定する場合に,大きく分けられることがわかった.また,標準画像に関しては,上記のような経緯,手順で作成されたにもかかわらず,他の研究者が,自身の研究を評価するために当初の目的外に流用し,それが学会などを通じて広まった結果,現在に至っているようである.つまり,画像研究の現場で一般に...
    文部科学省, 基盤研究(C), 東京工業大学, Coinvestigator not use grants, Competitive research funding, 18630002
  • Nonlinear Data Analysis for Functional Data
    Grants-in-Aid for Scientific Research(基盤研究(B))
    Apr. 2002 - Mar. 2005
    Masahiro MIZUTA; 鈴川 晶夫; 村井 哲也; 佐藤 義治; 小宮 由里子; 南 弘征
    In most conventional data analysis methods, we assume that data set is regarded as a set of numbers with some structures, for example a set of vectors or a set of matrices etc. Nowadays, we must often analyze more complex data. One type of the complex data is functional data structure ; data themselves are represented as functions. Ramsay and Silverman have studied function data analysis (FDA) as the analysis method to function data since the 1990's. They have published excellent books on FDA (Ramsay & Silverman, 1997, 2002).In this study, we promoted the research of the function data analy...
    Ministry of Education, Culture, Sports, Science and Technology, 基盤研究(B), 北海道大学, Coinvestigator not use grants, Competitive research funding, 14380122
  • 仮想並列計算機を活用したデータ解析法の開発と実装に関する研究
    科学研究費補助金(若手研究(B))
    2002 - 2003
    南 弘征
    本研究は,既存のパーソナルコンピュータをコンピュータネットワークで相互接続した仮想並列計算機環境において,並列計算機に適したデータ解析法・アルゴリズムを開発,実装するとともに,仮想並列計算機上であることをユーザに意識させることなく利用可能なグラフィカルインタフェースの構築を目的とする.平成15年度においては,平成14年度に得られた知見に基づき,並列計算機上で有効な,新たな解析手法を開発,検討するとともに,インターネットと親和性の高い,並列データ解析用のグラフィカルユーザインタフェースを構築することを目的とした.具体的な研究内容としては,平成14年度中に実装したブートストラップ法に加え,アルゴリズム中の部分計算の分割が容易であり,かつ,初期値が結果に及ぼす影響の大きいk-means法の並列化を行い,有効性を比較検討した.また,各々の代表的手法における仮想並列計算機内での挙動詳細を精査すべく,中途過程の履歴をGUIにより図示させ,アルゴリズムによる挙動の差異を確認した.その結果,代表的な探索的データ解析手法において,複数の部分計算の粒度が均一でない場合においても,MPIライブラリによる実装が安定した挙動を示すこと,PVMライブラリによる実装では,各ノードにおいて待ち時間が発生し,効率的計算という観点からは疑義が認められることなどが明らかになった.以上から,仮想並列計算機環境は既存...
    文部科学省, 若手研究(B), 北海道大学, Principal investigator, Competitive research funding, 14780165
  • インターネット上での分散データ解析システムに関する研究
    科学研究費補助金(奨励研究(A))
    1999 - 2000
    南 弘征
    交付申請書に記した通り、平成12年度は以下のような方針に則った形での、Java言語による既存システムの再構築を目標とした。1.解析作業の並列化を意図した複数プロセスの並行動作(スレッド)導入2.データ、解析手法、視覚的インタフェースのそれぞれに関する、オブジェクト指向に基づいたクラス構築3.解析部分のみ、あるいは視覚的環境のみを、流用可能な「部品」とするためのクラス構成、コーディングの調整うち、1および2,3の一部については成果を得、国際会議にて研究発表、意見交換を行なった。さらに、研究の進行過程において、主に項目1に関して子細に考察したところ、Java言語単独でのデータ解析システムの構築そのものは可能だが、Java仮想マシンの現状において実行速度の低下は不可避であり、例えば実習教材など、比較的軽微な解析を行なうには充分であるが、本格的な解析を行なうエンジンとしては適当ではない、との考察に至った。そこで、既存データ解析ソフトウェア以上のパフォーマンスを得るべく、また、Javaのスレッド機能を充分に活用することを視野に入れ、年度後半より、代表者所属先が有する1200台余のパーソナルコンピュータをTCP/IPネットワークにより仮想的に結合して並列計算機として動作させる、いわゆる「仮想並列計算機エンジン」を援用する形態のシステム開発にも取り組んだ。以上を要するに、(1)Javaによ...
    文部科学省, 奨励研究(A), 小樽商科大学->北海道大学, Principal investigator, Competitive research funding, 11780165
  • Nonlinear Data Analysis with Algebraic Curve Fitting
    Grants-in-Aid for Scientific Research(基盤研究(B))
    1997 - 2000
    Masahiro MIZUTA; 今井 英幸; 南 弘征; 鈴川 晶夫; 村井 哲也; 佐藤 義治
    The aim of the research is to construct a series of methods for nonlinear data analysis. Data analysis methods with algebraic curve fitting are fundamental methods in the research. We had studied algebraic curve fitting at first. We developed an algorithm to derive the exact distance between a point and algebraic curve. A method for finding an algebraic curve structures in data is proposed. According to the progress of the research, we also study Generalized Principal Components Analysis and Principal Curves. For related data analysis methods, Sliced Inverse Regression, Functional Data Anal...
    Ministry of Education, Culture, Sports, Science and Technology, 基盤研究(B), 北海道大学, Coinvestigator not use grants, Competitive research funding, 09480048
  • 複雑な標本調査におけるリサンプリング推定法に関する研究
    科学研究費助成事業 特定領域研究(A)
    1998 - 1998
    馬場 康維; 土屋 隆裕; 南 弘征
    1) 文献調査を行い,ブートストラップ法を分類・整理し,実用の面から評価した。
    2) 既存調査データ及び人工的なデータにより理論式の推定精度の評価とりサンプリング法による評価の比較をした。さらには,種々のサンプリングを想定したケースについて検討した。
    3) 大規模母集団からの標本調査にリサンプリング法が適用可能かどうか検討した。
    検討事項は,計算時間,扱えるデータ容量等,情報処理の観点からのものである。これらの検討の結果,リサンプリング法について下記の評価を得た。
    ・条件が整えば,精度の観点から,十分実用的である
    ・計算時間の観点から,問題はない
    ・計算機の容量の面から見ても,パーソナルコンピュータで扱える
    ・一般論的な理論面の検討がまだ不十分である
    これらの結果は,下記の学会等で報告した(1)。また,関連する研究の報告も行った。
    t衾歿貴
    (1) 馬場康維,土屋隆裕,ブートストラップ推定一理論と実用性,第66回日本統計学会,1998年7月
    (2) 南 弘征,統計科学におけるインターネットとの接点,統計数理研究所共同研究(A-38)による研究会報告,1998年10月
    (3) 南 弘征,水田正弘,視覚的データ解析システムにおけるクラス構成とJavaによる実装,第12回日本計算機統計学会シンポジウム,1998年11月
    日本学術振興会, 特定領域研究(A), 統計数理研究所, 10113216
  • インターネットを援用したデータ解析支援環境に関する研究
    科学研究費補助金(奨励研究(A))
    1997 - 1998
    南 弘征
    本研究は、加速度的に普及しているインターネットの技術基盤を用い、インターネットを通じてデータ解析処理手順を得ることができ、具体的処理の実行が個々のユーザの計算機環境下で行なわれるような、ネットワークを活用したデータ解析支援環境の構築を研究目標とした。うち、平成10年度は、平成9年度に得られた知見に基づきJavaによるシステムの再構成を主たる課題とした。しかし、研究期間中にJavaの基本開発環境であるJDK(Java Developers Kit)の版が再三にわたり更新され、特にVersion1.1から1.2に至る過程では解析支援環境に欠かせないポインティングデバイス(マウス)操作に係るインタフェース、プログラムモジュールの取り扱いが大幅に変更された。一応の互換性は考慮されていたものの、今後の開発を視野に入れ、ほぼ全面的に再構築を行なった。また、Javaを用いることにより当初想定していなかったグラフィクス活用を視野に入れる必然性が生じた。研究代表者は従来より視覚的データ解析環境についても研究を行なってきたところであり、グラフィクスと親和性の高い環境では、極力グラフィクスを援用することで、操作性は飛躍的に高まるとの知見を得ている。そこで、処理手順自体の支援ではなく、グラフィクスを介した形で処理列の妥当性を検証する方がJava環境には適していると判断した。以上の知見を踏まえたプロト...
    文部科学省, 奨励研究(A), 小樽商科大学, Principal investigator, Competitive research funding, 09780212
  • 複雑な標本調査におけるリサンプリング推定法に関する研究
    科学研究費助成事業 重点領域研究
    1997 - 1997
    馬場 康維; 土屋 隆裕; 南 弘征; 竹ヶ原 克彦
    1)複雑な標本調査の推定へのリサンプリング法の適用に関する研究の文献調査を行った。この調査の結果を分類・整理し,既存理論を実用の面から評価した。
    2)既存調査データ及び人工的なデータにより理論式の推定精度の評価とリサンプリング法による評価の比較をした。さらには,種々のサンプリングを想定したケースについて検討した。
    3)計算時間,扱えるデータ容量等,情報処理の観点からリサンプリング法の実用性について検討した。
    これらの検討の結果,リサンプリング法について下記の評価を得た。
    ・条件が整えば,精度の観点から,十分実用的である
    ・計算時間の観点から,問題はない
    ・計算機の容量の面から見ても,パーソナルコンピュータで扱える
    ・一般論的な理論面の検討がまだ不十分である
    これらの結果は,下記の学会等で報告した。
    学会報告
    1.馬場康維,土屋隆裕,中村好宏,小林良行,ブートストラップ法による標準誤差の推定の試み(2),第11回日本統計学会大会
    2.馬場康維,土屋隆裕,中村好宏,個票データの利用による推定精度の評価実験,第65回日本統計学会
    3.馬場康維,土屋隆裕,中村好宏,小林良行,労働力調査におけるブートストラップ法の利用
    日本学術振興会, 重点領域研究, 統計数理研究所, 09206218
  • RESEARCH CONCERNING EVALUATION METHOD AND PROCUREMENT PROCEDURE WITH MEDICAL EQUIPMENTS AND SERVICES IN INTERNATIONALIZATION SOCIETY
    Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
    1996 - 1997
    ABE Hiroshi; YAMAOKA Katsunori; MINAMI Hiroyuki; MITANI Kazufumi; SATOH Shoji; MATSUURA Tohru
    (1) RFPs and integrated scoring forms of the computers and the telecommunication equipments which had been procured by using OGVM in relation to the medical equipments were collected, digitalized, then the data table was made.
    (2) By the sentence structure analysis, dangerous factors which can invent misunderstandings and might cause the dispute the complaint from the suppliers as a result of the difference of the interpretation between the suppliers and procurement organizations have been extracted.
    (3) The RFP formats to control these factors were generated.
    (4) How setting the score, especially distribution theory of basic point and adding point, and how are a basic point and adding point related to the price and the performance of the proposal system were examined.
    (5) It was shown that the basic point part corresponded as a result inductively in the price and the part of adding point corresponded to the performance.
    (6) If the specific gravity in the basic point part was large, proper procurement became a price competition with discernment becoming to the performance competition when it in the part of adding point grew.
    (7) However, the existence of the problem as follows became clear.
    (a) The upper bound of the amount of money of procurement scale, that is, amount of the budget must not be shown.
    (b) The schedule price must be given as a constant number.
    (c) Mathematical theory of a basic point and adding point must be insufficient.
    (d) Proper evaluation technique concerning the characteristic evaluation object of OGVM,for instance, life cycle cost and the real operaion truth, has not become operation with a proper feretory in a present interpretation of an accounting method and a financial method. Therefore, the continuity examination of these problems to operate OVGM in the future by a more effective, steadier method and making to the theory are indispensable.
    Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B), HOKKAIDO UNIVERSITY, 08300006
  • Development of Visual Programming Environment for Exploratory Data Analysis
    Grants-in-Aid for Scientific Research(試験研究(B), 基盤研究(A))
    1995 - 1997
    Yoshiharu SATO; 南 弘征; 今井 英幸; 塩谷 浩之; 村井 哲也; 水田 正弘
    In computer science, "Visualization" is one of the most attractive keywords. "Visualization" is useful not only to reveal data but to deal with analysis processes. Statisticians are often forced to apply many methods to the same data in their analysis since they wants to find the most reasonable result. A lot of results they made, however, may perplex them. If they can view an analysis flow (made of some statistical methods) on the graphical display, they can easily find the relations between data and applied methods. It is called "Visualization of Execution."As we described, "Visualization...
    Ministry of Education, Culture, Sports, Science and Technology, 試験研究(B), 基盤研究(A), 北海道大学, Coinvestigator not use grants, Competitive research funding, 07558149
  • 知識工学を応用したデータ解析支援システムの開発
    科学研究費補助金(奨励研究(A))
    1995 - 1995
    南 弘征
    本研究では、統計学に関する知識をあまり持たない利用者に対して、適切な解析作業・解析結果を与えるために、知識工学を応用して計算機側からの支援を行なうシステムの構築を目指した。具体的には、研究代表者により本研究以前に作成された試作システムに対し、以下に述べる3点の改良を試みた。まず、アルゴリズム再検討による推論速度の改良については、非単調推論アルゴリズムの導入により、推論過程そのものには改善がみられ、統計家の行なう判断に近付いたものの、取り扱わねばならない中途過程、述語数が増大し、知識工学におけるいわゆるフレーム問題と直面することとなった。研究前の予想を覆すほどの量であり、述語数の取捨選択では対処しきれないものと思われ、目的自体は半ば達成したものの、今後、知識ベースに関する再検討の必要性も研究結果として得られたことになる。次に、判断内容再検討による適用範囲拡大については、先に述べた通り、取り扱う知識量の増大により、わずかに数種の新手法、判断基準を盛り込むに留まった。知識として1基準増えるだけでも、推論過程では指数的に増大してしまうためである。また、グラフィクスを用いたインタフェース改良については、当初要求していた研究費が得られず、開発環境として不充分な状態であったが、幸い、研究代表者が研究分担者として参加している試験研究(B)(2)「探索的データ解析のための視覚的プログラミング環...
    文部科学省, 奨励研究(A), 小樽商科大学, Principal investigator, Competitive research funding, 07780198