A proposal of structural mixture distribution model. its application and basic analysis-

2011 Fiscal Year Final Research Report

• Project/Area Number: 21500146
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Keio University
• Principal Investigator: SAKURAI Akito 慶應義塾大学, 理工学部, 教授 (00303339)
• Project Period (FY): 2009 – 2011
• Keywords: 知的システムアーキテクチャ

Research Abstract

For the structural mixture model as a learning model, new methods to estimate mixture coefficients were proposed. It outperformed or was equivalent in performance to existing methods on real datasets evaluation. On price formation in financial markets which are considered efficient, existence of an element model was indicated by simulation trading in the market.

Research Products

• [Presentation] Prediction of Foreign Exchange Market States with Support Vector Machine (2011)
  • Author(s): Kei Shioda, Deng Shangkun, and Sakurai Akito
  • Organizer: The Tenth International Conference on Machine Learning and Applications 2011
  • Place of Presentation: Honolulu, Hawaii
  • Year and Date: 2011-12-20
• [Presentation] Multiple Kernel Learning on Time Series Data and Social Networks for Stock Price Prediction (2011)
  • Author(s): Deng Shangkun and Sakurai Akito
  • Organizer: ICMLA 2011 Special Session on Learning on the Web
  • Place of Presentation: Honolulu, Hawaii
  • Year and Date: 2011-12-19
• [Presentation] Combining Technical Analysis with Sentiment Analysis for Stock Price Prediction (2011)
  • Author(s): Deng Shangkun and Sakurai Akito
  • Organizer: International Conference on Social Computing and its Applicaitons(SAC 2011)
  • Place of Presentation: Sydney, Australia
  • Year and Date: 2011-12-13
• [Presentation] Manifold-Regularized Minimax Probability Machine (2011)
  • Author(s): K Yoshiyama and A Sakurai
  • Organizer: IAPR Workshop on Partially Supervised Learning(PSL 2011)
  • Place of Presentation: Universitat Ulm, Germany
  • Year and Date: 2011-09-16
• [Presentation] Combining Multiple Kernel Learning and Genetic Algorithm for Forecasting Short Time Foreign Exchange Rate (2011)
  • Author(s): Deng Shangkun and Sakurai Akito
  • Organizer: The Eleventh IASTED International Conference on Artificial Intelligence and Applications AIA 2011
  • Place of Presentation: Innsbruck, Austria
  • Year and Date: 2011-02-15
• [Book] Self-Organization and Aggregation of Undisclosed Knowledge. In Self Organizing Maps-Applications and Novel Algorithm Design (2011)
  • Author(s): K Ishikawa, Y Shinozawa, and A Sakurai
  • Total Pages: 143-172
  • Publisher: INTECH