Development and application of next generation pattern recognition methods for user behavior analysis using large scale log data

2016 Fiscal Year Final Research Report

• Project/Area Number: 26560167
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Social systems engineering/Safety system
• Research Institution: Waseda University
• Principal Investigator: GOTO Masayuki 早稲田大学, 理工学術院, 教授 (40287967)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 大規模データ / 大規模ログデータ / パターン認識 / 機械学習 / ユーザ行動 / 潜在クラスモデル / テキストマイニング / 時系列分析

Outline of Final Research Achievements

The objective of this study is to develop next-generation pattern recognition methods for management decision making and marketing technology through the analysis of user behaviors based on large-scale log data accumulated in databases such as EC sites. While analyzing actual user behavior history data, we studied new models and methods which are theoretically considered highly versatile. In particular, we developed a machine learning model for developing the issuing logic for real-time coupon ticket systems from multiple viewpoints on users' page browsing behaviors on an EC site. The performances of our proposed models and analytical techniques were verified by using actual data and various applications were demonstrated.

Free Research Field

経営情報工学