Establishment of Co-occurring Cluster Mining and Its Environmental Contribution

2014 Fiscal Year Final Research Report

• Project/Area Number: 24650068
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Intelligent informatics
• Research Institution: Osaka University
• Principal Investigator: KEN-ICHI Fukui 大阪大学, 産業科学研究所, 助教 (80418772)
• Co-Investigator(Renkei-kenkyūsha): 佐藤 一永 東北大学 (50422077) ; 水崎 純一郎 東北大学 (90092345) ; 沼尾 正行 大阪大学 (30198551)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: データマイニング / クラスタリング / 共起 / 環境貢献 / 燃料電池 / 地震 / 相互作用

Outline of Final Research Achievements

In this research, as an objective is to infer interaction between events, we developed a novel mining algorithm that extracts pairs of clusters in which events co-occur, from a observed event sequence data. Firstly, by using simulated data, we validated that the proposed method can extract co-occurring clusters more accurately compared with a two-step method. Next, we applied the proposed method to damage analysis on a fuel cell, and confirmed that our method can extract valid mechanical interactions among members of the fuel cell from Acoustic Emission event sequence. Lastly, we also applied the method to analysis of Earthquake occurrences after the 2011 Tohoku Earthquake, and the results suggested interactions among asperities.

Free Research Field

データマイニング