An immediate analysis and automatic tag extraction for sensor databases

2016 Fiscal Year Final Research Report

• Project/Area Number: 26330141
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Multimedia database
• Research Institution: Komazawa University
• Principal Investigator: Yoshida Naofumi 駒澤大学, グローバル・メディア・スタディーズ学部, 准教授 (10338238)
• Research Collaborator: Honda Ken 駒澤大学, 大学院 グローバル・メディア研究科
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: マルチメディア・データベース / データモデル / センサーデータ / データ分析

Outline of Final Research Achievements

In this research project, the principle aim is an automatic extraction of tags for sensor databases by immediate analysis of data. The system analyses sensor data from databases immediately, and outputs the results as tags. Based on the mutual resource exchanging model (our previous work), we realized an three dimensional movie generation method by collective intelligence, and an hybrid learning method by mutual exchanging learning results. Also, we have presented a credibility evaluation method by integrating legacy databases and sensor databases. We have shown these results in journal papers and international conference papers.

Free Research Field

データベースシステム，マルチメディアシステム