多種情報源からのデータマイニング手法による肝炎知識の発見

2006 Fiscal Year Annual Research Report

• Project/Area Number: 16300043
• Research Institution: Japan Advanced Institute of Science and Technology
• Principal Investigator: TU・BAO Ho 北陸先端科学技術大学院大学, 知識科学研究科, 教授 (60301199)
• Co-Investigator(Kenkyū-buntansha): 高林 克日己 千葉大学, 医学部附属病院, 教授 (90188079) ; 河崎 さおり 北陸先端科学技術大学院大学, 知識科学研究科, 助手 (40377437)
• Keywords: 肝炎 / データマイニング / 多種情報源 / 時系列抽象化 / 医学文献 / 専門家の評価 / 統合的アプローチ

Research Abstract

多様な情報源を相互活用する知識発見手法を提案し、診療データから肝炎に関する有用な知識を発見することを目的とした本研究は、この研究期間に以下の実績をあげ、当初設定した研究目的をほぼ達成することができた。
1.時区間関係を活用する時相論理に基づく時系列抽象化の新手法を開発した。この新手法は、診療記録の時系列データを解析し、データ中のありうる全時区間関係を抽出するものであり、えられた関係を用いて、肝炎研究の上の問題に関する規則を発見した。
2.巨大なテキスト情報源を活用するためのテキストマイニング手法(条件付確率場、トレランス・ラフ集合などの先端の機械学習技法を用いたもの)を開発し、(a)学習段階の探索空間を狭めるための背景知識、(b)診療データに対するデータマイニングで発見した規則を解釈するための知識という二通りの知識発見を目的としてMEDLINEに適用した。
3.学習手法、医学文献、専門家による判断を相互に組合せる医療データマイニングへの統合アプローチを開発し、単にそれぞれの手法を用いるよりもこのアプローチの方が効率的であることを示した。
4.その他の成果としては、基礎研究においてカーネル手法、データ・発見プロセスおよび規則の可視化、ルール帰納法、最適化の諸課題に、バイオインフォマティクス領域でタンパク質間相互作用の問題に取り組み、その成果を、先端的な学習手法を用いた医療データ研究の継続および、診療記録データと病原遺伝子との関係の解明という方向につなげることができた。

Research Products

• [Journal Article] A hyper-heuristic for descriptive rule induction (2007)
  • Author(s): Pham, T.H., Ho, T.B.
  • Journal Title: International Journal of Data Warehousing and Mining 3-1 Pages: 54-66
  • Description: 「研究成果報告書概要(和文)」より
• [Journal Article] Integration of learning methods, medical literature and expert inspection in medical data mining (2007)
  • Author(s): Ho, T.B., Kawasaki, S., Takabayashi, K., Nguyen, C.H.
  • Journal Title: IEICE Trans.Information and Systems (In press)
• [Journal Article] Temporal Relations Extraction in Mining Hepatitis Data (2007)
  • Author(s): Ho, T.B., Nguyen, C.H., Kawasaki, S., Le, S.O., Takabayashi, K.
  • Journal Title: Journal of New Generation Computing 25-3(In press)
• [Journal Article] Temporal Abstraction and Data Mining with Visualization of Laboratory Data (2007)
  • Author(s): Takabayashi, T., Ho, T.B., Yokoi, H., Nguyen, T.D., Kawasaki, S., Le, S.Q., Suzuki, T., Yokosuka, O
  • Journal Title: The Proceedings of Medinfo 2007 Congress, 20-24 August, Brisbane (In press)
• [Journal Article] Kernel Matrix Evaluation (2007)
  • Author(s): Nguyen, C.H., Ho, T.B.
  • Journal Title: The proceedings of Twentieth International Joint Conference on Artificial Intelligence(IJCAI'07),January 6-12,Hyderabad, India Pages: 987-992
• [Journal Article] Wrapper feature extraction for time series classification using singular value decomposition (2006)
  • Author(s): Zhang, H., Ho, T.B., Kawasaki, S.
  • Journal Title: International Journal of Knowledge and Systems Science 3-1 Pages: 53-60
  • Description: 「研究成果報告書概要(和文)」より
• [Journal Article] A Bottom-up Method for Simplifying Support Vector Solutions (2006)
  • Author(s): Nguyen, D.D., Ho, T.B.
  • Journal Title: IEEE Transactions on Neural Networks 17-3 Pages: 792-796
  • Description: 「研究成果報告書概要(和文)」より
• [Journal Article] Asynchronous Constraint-based Approach - New Solution for Any Constraint Problem, Workshop Rational, Robust, and Secure Negotiations in Multi-Agent Systems (2006)
  • Author(s): Ben Hassine, A., Ho, T.B.
  • Journal Title: The proceedings of The Fourth International Joint Conference on Autonomous Agents and Multi Agent Systems(AAMAS'06),Hakodate
• [Journal Article] Scientific data mining : Opportunities and challenges(invited talk) (2006)
  • Author(s): Ho, T.B.
  • Journal Title: The proceedings of International Symposium on Towards Creating New Industries Based on Inter-Nanoscience, December 8-9,Awaji, Hyogo Pages: 61-64
• [Journal Article] Sampling for Imbalanced Data Learning (2006)
  • Author(s): Nguyen, C.H., Ho, T.B.
  • Journal Title: The proceedings of The International Workshop on Data-Mining and Statistical Science(DMSS2006),September 25-26,Sapporo Pages: 12-19
• [Journal Article] Parallel Training of CRFs : A Practical Approach to Build Large-Scale Prediction Models for Sequence Data (2006)
  • Author(s): Phan, X.H., Nguyen, L.M., Horiguchi, S., Inoguchi, Y., Ho, T.B.
  • Journal Title: The proceedings of Workshop on Parallel Data Mining, ECML/PKDD 2006,Berlin, September 18-22 Pages: 51-63
• [Journal Article] Feature Extraction for Time Series Classification Using Discriminating Wavelet Coefficients (2006)
  • Author(s): Zhang, H., Ho, T.B., Lin, M.S., Liang, X
  • Journal Title: The proceedings of Third International Symposium on Neural Networks(ISNN 2006),28-31 May, Chengdu, China(Lecture Notes in Artificial Intelligence LNAI 3971,Springer) Pages: 1394-1399
• [Journal Article] Multiscale Feature Extraction for Time Series Classification with Hybrid Feature Selection (2006)
  • Author(s): Zhang, H., Lin, S.M., Huang, W., Kawasaki, S., Ho, T.B.
  • Journal Title: The proceedings of International Conference on Intelligent Computing(ICIC'06),August 16-19,Kunming, China(Lecture Notes in Control and Information Sciences LNCIS 344) Pages: 939-944
• [Journal Article] An Integrated Approach in Medical Data Mining, The First International Conference on Knowledge (2006)
  • Author(s): Kawasaki, S., Ho, T.B.
  • Journal Title: The proceedings of The First International Conference on Knowledge, Information and Creativity Support Systems(KICSS'06),1-4 August, Ayutthaya, Thailand Pages: 24-31
• [Journal Article] Knowledge Visualization in Hepatitis Study (2006)
  • Author(s): Nguyen, D.D., Ho, T.B., Kawasaki, S.
  • Journal Title: The proceedings of Asia Pacific Symposium on Information Visualization(APVIS 2006),February 1-3,Tokyo Pages: 59-62
• [Journal Article] Discovering signal transduction networks using signaling domain-domain interactions (2006)
  • Author(s): Nguyen, T.P., Ho, T.B.
  • Journal Title: The proceedings of Genome Informatics 2006,December 18-20,Yokohama Pages: 35-45
• [Journal Article] Association-Based Dissimilarity Measures for Categorical Data : Limitation and Improvement (2006)
  • Author(s): Le, S.Q., Ho, T.B., Le, S.V.
  • Journal Title: 10th Pacific-Asia Conference on Knowledge Discovery and Data Mining PAKDD'06,9-12 April, Singapore.Lecture Notes in Artificial Intelligence LNAI 3918 Pages: 493-498
• [Journal Article] Prediction of domain-domain interactions using inductive logic programming from multiple genome databases (2006)
  • Author(s): Nguyen, T.P., Ho, T.B.
  • Journal Title: Springer LNAI 4256 : The proceedings of Discovery Science(DS'06),7-10 October, Barcelona, Spain Pages: 185-196
• [Journal Article] Exploiting Temporal Relations in Mining Hepatitis Data (2006)
  • Author(s): Takabayashi, K., Ho, T.B., Nguyen, C.H., Kawasaki, S., Yokoi, H., Suzuki, T., Yokosuka, O.
  • Journal Title: The proceedings of he 26th Joint Conference on Medical Informatics, November 1-4,Sapporo
• [Journal Article] Conditional Random Fields for Predicting and Analyzing Histone Occupancy, Acetylation and Methylation Areas in DNA Sequences (2006)
  • Author(s): Tran, D.H., Pham, T.H., Satou, K., Ho, T.B.
  • Journal Title: The proceedings of 4th European Workshop on Evolutionary Computation and Machine Learning in Bioinformatics, Budapest,10-12 April Pages: 221-230
• [Journal Article] Combining the Global and Partial Information for Distance-Based Time Series Classification and Clustering (2005)
  • Author(s): Zhang, H., Ho, T.B., Lin, M.S., Huang, W.
  • Journal Title: Journal of Advanced Computational Intelligence ＆ Intelligent Informatics 10-1 Pages: 69-76
  • Description: 「研究成果報告書概要(和文)」より
• [Journal Article] Springer-Verlag
  • Author(s): Huynh, V.N., Nakamori, Y., M.Ryoke, Ho, T.B.
  • Journal Title: Fuzzy Optimization and Decision Making(Decision making under uncertainty with fuzzy targets)
• [Book] Creative Environments(Chapter 4,Knowledge Acquisition by Machine Learning and Data Mining) (2007)
  • Author(s): Ho, T.B., Kawasaki, S., Granat, J.
  • Publisher: Springer-Verlag(In press)
• [Book] Fuzzy Optimization and Decision Making(Decision making under uncertainty with fuzzy targets) (2007)
  • Author(s): Huynh, V.N., Nakamori, Y., M.Ryoke, Ho, T.B.
  • Publisher: Springer-Verlag(In press)
• [Book] Advances in Data Warehousing and Mining (2007)
  • Author(s): Ho, T.B., Nguyen, T.P., Tran, T.N.
  • Publisher: IGC Publishers(In press)
• [Book] Theories and Applications of Computer Science Advances in Knowledge Discovery and Data Mining (2006)
  • Author(s): Ho, T.B., Dong, T.B.T., Duong, A.D., Nguyen, D.T
  • Total Pages: 260
  • Publisher: World Scientific