Mining algorithm for sequential patterns with multiple concepts of time

2017 Fiscal Year Final Research Report

• Project/Area Number: 26330129
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Multimedia database
• Research Institution: The University of Electro-Communications
• Principal Investigator: Shintani Takahiko 電気通信大学, 大学院情報理工学研究科, 准教授 (30604623)
• Project Period (FY): 2014-04-01 – 2018-03-31
• Keywords: データマイニング / シーケンスデータ / 時間概念 / 長時間パターン

Outline of Final Research Achievements

This research work proposed a method for finding sequential patterns containing various time concepts from sequence data of data events with duration. We investigated the time concepts and patterns by real application and clarified that it is important to evaluate patterns by pattern duration instead of frequency. We then defined a long duration pattern that occupied a long time, extended to a long duration pattern containing various time concepts by considering the duration and start time of data events, and proposed an efficient method to find long duration patterns by omitting redundant counts. We confirmed the effectiveness of our method by experiments using real data.

Free Research Field

データ工学