構造データからのアクティブマイニング

• 研究課題/領域番号: 13131206
• 研究種目: 特定領域研究
• 配分区分: 補助金
• 審査区分: 理工系
• 研究機関: 大阪大学
• 研究代表者: 元田 浩 大阪大学, 産業科学研究所, 教授 (00283804)
• 研究分担者: 鷲尾 隆 大阪大学, 産業科学研究所, 助教授 (00192815) ; 大原 剛三 大阪大学, 産業科学研究所, 助手 (30294127) ; TUBAO Ho (HO Tu Bao) 北陸先端科学技術大学院大学, 知識科学研究科, 教授 (60301199) ; 矢田 勝俊 関西大学, 商学部, 助教授 (00298811) ; 吉田 哲也 北海道大学, 大学院・情報科学研究科, 助教授 (80294164)
• 研究期間 (年度): 2001 – 2004
• 研究課題ステータス: 完了 (2004年度)
• 配分額: 64,800千円 (直接経費: 64,800千円); 2004年度: 13,400千円 (直接経費: 13,400千円); 2003年度: 16,400千円 (直接経費: 16,400千円); 2002年度: 16,800千円 (直接経費: 16,800千円); 2001年度: 18,200千円 (直接経費: 18,200千円)
• キーワード: 機械学習 / 知識発見 / データマイニング / クラスタリング / 視覚化 / 時系列解析 / データ前処理 / 時系列相関

研究概要

研究実績は以下のとおり.
1.グラフ構造データからの決定木構築プログラムDT-GBIでの探索過程に領域知識の制約(指定したパターンを含む,含まない)を導入した.
2.ペアを仮想ノードとして扱い,チャンキングをせず探索する新グラフマイニング手法Cl-GBIを開発した.適切なパラメータ設定により完全探索が可能になり,GBIの数え落としの問題点などを解決した.
3.上記Cl-GBIを組み込んだ決定木構築プログラムDT-ClGBIを開発し,肝炎データセットで性能を評価した.
4.数値データを伴うデータから,数値を記号離散化することなしに相関の高い数値区間を自動抽出する原理を確立し,それに基づく数値相関規則導出手法を開発した.
5.ユーザ指向データマイニングシステムD2MSの肝炎患者に関するルールの理解容易性向上を確認し,多数のルールから統計的に有意なものを選定する手法とルール学習において領域知識を表現の制約に加える手法を提案した,
6.科学データマイニングとしてゲノムおよび結晶データを並行して解析した.前者に関しては,SVMによるタンパク質の2次構造におけるβターンの予測手法を拡張しγターンを予測した.後者に関しては,粉末回折データから結晶構造を同定する手法を遺伝的アルゴリズムに基づき開発した.
7.意味的まとまりを捉えたパッセージの集合として文書を表し,トレランス・ラフ集合モデルによるソフトマッチを導入し,意味を反映した相関ルールを得る手法を開発した.
8.グラフマイニング手法AGMを消費者行動データに適用し,アクティブマイニングによる実証実験を行い新しいデータを収集した.アルコール市場分析から得られた知見に基づき,実際の店舗で店頭プロモーションを行った結果,対象商品の売り上げ増加,関連商品の同時購入頻度の増加を検証することができた.

研究成果

• [雑誌論文] Enhancing the plausibility of law equation discovery through cross check among multiple scale-type-based models (2005)
  • 著者名/発表者名: T.Washio
  • 雑誌名: Journal of Experimental & Theoretical Artificial Intelligence Vol.17/1-2 ページ: 129-143
• [雑誌論文] 多頻度マイニング手法の一般化 (2004)
  • 著者名/発表者名: 猪口 明博
  • 雑誌名: 人工知能学会論文誌 Vol.19/5 ページ: 368-378
• [雑誌論文] A selective sampling approach to active feature selection (2004)
  • 著者名/発表者名: H.Liu
  • 雑誌名: Artificial Intelligence Vol.159/1-2 ページ: 49-74
• [雑誌論文] Preliminary Analysis of Interferon Therapy by Graph-Based Induction (2004)
  • 著者名/発表者名: T.Yoshida
  • 雑誌名: Proc.of the Third International Workshop on Active Mining (AM-2004) Vol.3/1 ページ: 31-40
• [雑誌論文] Application of DT-GBI to Promoter and Hepatitis Datasets (2004)
  • 著者名/発表者名: H.Motoda
  • 雑誌名: Proc.of the Knowledge Discovery in BioMedicine Vol.1/1 ページ: 10-40
• [雑誌論文] Constructing Compact Dual Ensembles for Efficient Active Learning (2004)
  • 著者名/発表者名: A.Mandvika
  • 雑誌名: Proc.of the Pacific Knowledge Acquisition Workshop Vol.5/1 ページ: 29-43
• [雑誌論文] Using a Hash-based Method for Apriori-based Graph Mining (2004)
  • 著者名/発表者名: P.C.Nguyen
  • 雑誌名: Proc.of the 8th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD) LNAI-3202 ページ: 349-361
• [雑誌論文] Compact Dual Ensembles for Active Learning (2004)
  • 著者名/発表者名: A.Mandvika
  • 雑誌名: Proc.of 8th Pacific-Asia Conference on Knowledge Discovery and Data Mining LNAI-3056 ページ: 293-297
• [雑誌論文] Analysis of Hepatitis Dataset by Decision Tree Graph-Based Induction (2004)
  • 著者名/発表者名: K.Ohara
  • 雑誌名: Prof. of Discovery Challenge Workshop held in conjunction with PKDD-04 Vol.DC04/1 ページ: 173-184
• [雑誌論文] Constructive Inductive Learning based on Meta-Attributes (2004)
  • 著者名/発表者名: K.Ohara
  • 雑誌名: Proc.of the 7th International Conference on Discovery Science LNAI-3245 ページ: 142-154
• [雑誌論文] Consumer Behavior Analysis by Graph Mining Technique (2004)
  • 著者名/発表者名: K.Yada
  • 雑誌名: Proc.of the 8th International Conference on Knowledge-Based Intelligent Information and Engineering Systems LNAI-3214 ページ: 800-806
• [雑誌論文] Cl-GBI : A Novel Strategy to Extract Typical Patterns from Graph Data (2004)
  • 著者名/発表者名: P.C.Nguyen
  • 雑誌名: Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ and IEICE-SIGAI on Active Mining KBS-A403 ページ: 105-110
• [雑誌論文] Example-Based Reduction Using Hidden Markov Model (2004)
  • 著者名/発表者名: L.M.Nguyen
  • 雑誌名: ACM Transactions on Asian Language Information Processing Vol.3/3 ページ: 146-158
• [雑誌論文] Mining yeast transcriptional regulatory modules from factor DNA-binding sites and gene expression data (2004)
  • 著者名/発表者名: T.H.Pham
  • 雑誌名: Genome Informatics Vol.14/2 ページ: 287-295
• [雑誌論文] A novel graph-based similarity measure for 2D chemical structures (2004)
  • 著者名/発表者名: S.Q.Le
  • 雑誌名: Genome Informatics Vol.14/2 ページ: 82-91
• [雑誌論文] A Context Model for Constructing Membership Functions of Fuzzy Concepts Based on Modal Logic (2004)
  • 著者名/発表者名: V.N.Huynh
  • 雑誌名: Journal of Information Sciences Vol.160/1-4 ページ: 111-129
• [雑誌論文] Crystal Structure Determination using Data Mining Methods (2004)
  • 著者名/発表者名: T.B.Ho
  • 雑誌名: Proc.of the 5th International Symposium on Knowledge and Systems Sciences Vol.5/1 ページ: 81-86
• [雑誌論文] Detecting Emerging Trends from Scientific Corpora (2004)
  • 著者名/発表者名: M.H.Le
  • 雑誌名: Proc.of the 5th International Symposium on Knowledge and Systems Sciences Vol.5/1 ページ: 243-248
• [文献書誌] A.Inokuchi: "Complete Mining of Frequent Patterns from Graphs : Mining Graph Data"Machine Learning. Vol.50, No.3. 321-354 (2003)
• [文献書誌] 西村芳男: "AGMアルゴリズムの高速化と立体構造解析への適用"人工知能学会論文誌. Vol.18, No.5C. 257-268 (2003)
• [文献書誌] S.Kawasaki: "Temporal Abstraction for Long-Term Changed Tasts in the Hepatitis Domain"Journal of Advanced Computational Intelligence & Intelligent Informatics. Vol.170, No.3. 348-354 (2003)
• [文献書誌] K.Terai: "Business Process Semi-Automation Based on Business Model Management"International Journal of Intelligent Systems in Accounting, Finance and Management. Vol.11, No.4. 215-234 (2003)
• [文献書誌] T.B.Ho: "A Knowledge Discovery System with Support for Model Selection and Visualization"Applied Intelligence, Kluwer Academic Publishers. Vol.19, No.1-2. 125-141 (2003)
• [文献書誌] T.B.Ho: "Chance Discovery and Learning Minority Classes"Journal of New Generation Computing. Vol.21, No.2. 147-160 (2003)
• [文献書誌] H.Liu: "Active Feature Selection Using Classes"Proceedings of the 7th Pacific-Asia Conference on Knowledge Discovery and Data Mining. LNAI2637. 474-485 (2003)
• [文献書誌] W.Geamsakul: "Constructing a Decision Tree for Graph Structured Data"Proceedings of the First International Workshop on Mining Graphs, Trees and Sequences (MGTS-2003). 1-10 (2003)
• [文献書誌] A.Inokuchi: "Specific Biasis for Mining Frequent Substructures"Proceedings of the First International Workshop on Mining Graphs, Trees and Sequences (MGTS-2003). 45-54 (2003)
• [文献書誌] F.Adachi: "Development of Generic Search Method Based on Transformation Invariance"Proceedings of the 14th International Symposium on Methodologies for Intelligent Systems (ISMIS03). 486-495 (2003)
• [文献書誌] T.D.Nguyen: "Discovery of Trends and States in Irregular Temporal Data Abstraction of Long-Term Changed Attributes in the Hepatitis Domain"Proceedings of the 6th International Conference on Discovery Science. 410-417 (2003)
• [文献書誌] T.B.Ho: "Mining Hepatitis Data with Temporal Abstraction"Proceedings of ACM International Conference on Knowledge Discovery and Data Mining KDD-03. 369-377 (2003)
• [文献書誌] S.Kawasaki: "Abstraction of Long-Term Changed Attributes in the Hepatitis Domain"Proceedings of the 7th International Conference on Knowledge-based Intelligent Information and Intelligent Informatics. 365-372 (2003)
• [文献書誌] P.C.Nguyen: "Temporal Decomposition : A Promising Approach to VQ-based Speaker Identification"Proceedings of Intenational Fuzzy Systems Association World Congress. 184-187 (2003)
• [文献書誌] K.Yada: "Knowledge Discovery Process from Sales Data"Proceedings of the 2003 IRMA International Conference. 161-187 (2003)
• [文献書誌] T.Araki: "Chance Discovery from Consumer Research Using Key Graph"Proceedings of A Full-day Workshop in HCI International 2003. 45-54 (2003)
• [文献書誌] N.Katoh: "Business Application for Sales Transaction Data by Using Genome Analyis Technology"Proceedings of the 6th International Conference of Discovery Science. 208-219 (2003)
• [文献書誌] K.Yada: "Data Mining Oriented CRM Systems Based on MUSASHI : C-MUSASHI"Proceedings of the Second International Workshop on Active Mining. 52-61 (2003)
• [文献書誌] T.Matsuda: "Graph-Based Induction and Its Applications, Advanced Engineering Informatics"Advanced Engineering Informatics. Vol.16, No.2. 135-143 (2002)
• [文献書誌] H.Motoda: "Feature Selection, Extraction and Construction"Proc. of Foundation of Data Mining, PAKDD02. 67-72 (2002)
• [文献書誌] T.Matsuda: "Knowledge Discovery from Structured Data by Beam-wise Graph-Based Induction"Proc. of the Seventh Pacific Rim International Conference on Artificial Intelligence. 255-264 (2002)
• [文献書誌] T.Matsuda: "Beam-wise Graph-Based Induction for Structured Data Mining"Proc. of Pacific Rim Knowledge Acquisition Workshop (PKAW2002). 174-188 (2002)
• [文献書誌] T.Matsuda: "Mining Patterns from Structured Data by Beam-wise Graph-Based Induction"Induction, Workshop Notes on Multi-Relational Data Mining, KDD-2002. 113-129 (2002)
• [文献書誌] T.Matsuda: "Preliminary Analysis of Hepatitis Data by Beam-wise Graph-Based Induction"Discovery Challenge Workshop Notes, ECML/PKDD 2002, ISBN 952-10-0639-0. 60-72 (2002)
• [文献書誌] T.Matsuda: "Active Mining from Hepatitis Data by Beam-wise GBI"Proceedings of International Workshop on Active Mining. 37-44 (2002)
• [文献書誌] A.Inokuchi: "General Freamework for Mining Frequent Patterns in Structure"Proceedings of International Workshop on Active Mining. 23-30 (2002)
• [文献書誌] T.B.Ho: "Nonhierarchical Document Clustering by a Tolerance Rough Set Model"International Journal of Intelligent Systems. Vol.17, No.2. 199-212 (2002)
• [文献書誌] T.B.Ho: "Cluster-based Information Retrieval with a Tolerance Rough Set Model"International Journal of Fuzzy Logic and Intelligent Systems. Vol.2, No.1. 26-32 (2002)
• [文献書誌] T.B.Ho: "Chance Discovery and Learning Minority Classes"Journal of New Generation Computing, Special Issue on Chance Discovery. Vol.21, No.2. 147-160 (2003)
• [文献書誌] T.B.Ho: "Visualization Support for a User-Centered KDD Process"ACM International Conference on Knowledge Discovery and Data Mining, KDD-02, Edmonton. 519-524 (2002)
• [文献書誌] D.D.Nguyen: "Learning Minority Classes in Unbalanced Datasets"Third International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT'02). 196-203 (2002)
• [文献書誌] T.B.Ho: "Extracting Knowledge from Hepatitis Data with Temporal Abstraction"Discovery Challenge Workshop Notes, ECML/PKDD 2002, ISBN952-10-0639-0. 40-50 (2002)
• [文献書誌] K.Yada: "The Future Direction of Active Mining in the Business World"Frontiers in Artificial Intelligence and Applications. Vol.79. 239-245 (2002)
• [文献書誌] Y.Hamuro: "Discovering Purchase Association among Brands from Purchase History"Proceedings CD of SSGRR2002w. 67-77 (2002)
• [文献書誌] Y.Hamuro: "MUSASHI : Flexible and Efficient Data Preprocessing Tool for KDD based on XML"DCAP2002 Workshop held in conjunction with ICDM2002. 38-49 (2002)
• [文献書誌] M.Terabe: "Attribute Generation Based on Association Rules"Knowledge and Information Systems, An International Journal. Vol.4, No.3. 329-349 (2002)
• [文献書誌] 鈴木篤之: "システムの設計・運用・評価,岩波講座現代工学の基礎"岩波書店. 165 (2002)
• [文献書誌] 松田 喬: "一般グラフ構造データに対するGraph-Based Inductionとその応用"人工知能学会論文誌. 16. 363-374 (2001)
• [文献書誌] Takuya Wada: "Knowledge Acquisition from Both Human Expert and Data"Proc. of the Fifth Pacific-Asia Conference on Knowledge Discovery and Data Mining. 550-561 (2001)
• [文献書誌] Tu Bao Ho: "Visualization Support for User-Centered Model Selection in Knowledge Discovery and Data Mining"International Journal of Artificial Intelligence Tools. 691-713 (2001)
• [文献書誌] 矢田 勝俊: "経営データからの知識発見"国民経済雑誌. 184. 19-33 (2001)
• [文献書誌] 寺邊 正大: "S^3Baggingによる高速な分類器生成"数理モデル化と応用. 42. 25-38 (2001)
• [文献書誌] Tu Bao Ho: "Cluster-based Information Retrieval with Tolerance Rough Set Model"Proc. of Second International Symposium on Advanced Intelligent Systems. 6-11 (2001)
• [文献書誌] Huan Liu: "Instance Selection and Construction for Data Mining"Kluwer Academic Publishers. 416 (2002)