Speeding up the similarity search by generalized pivots

2016 Fiscal Year Final Research Report

• Project/Area Number: 26330138
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Multimedia database
• Research Institution: University of Shizuoka
• Principal Investigator: IKEDA Tetsuo 静岡県立大学, 経営情報学部, 教授 (60363727)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 情報検索 / 類似検索 / 可視化 / クラスタリング

Outline of Final Research Achievements

The purpose of our research is to develop efficient similarity search methods of multimedia data such as picture images and movies. (1)We proposed a pivot generation method characterized in that each pivot vector element can be improved independently and at high speed, and confirmed its effectiveness using L1 distance. (2)As a data visualization method, we proposed a method to extract features of partial data sets as annotations based on Z scores of attribute distributions, and confirmed its effectiveness. (3)As a clustering method, we proposed a method characterized by constructing pivots in the preprocessing process and then generating clusters, and confirmed its effectiveness.
They are technologies having originality with respect to similarity search and related large amount data utilization.

Free Research Field

データ工学