Discovery of knowledge about implicit relationships utilizing semantics, web data, and statistics

2018 Fiscal Year Final Research Report

• Project/Area Number: 15K00423
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Web informatics, Service informatics
• Research Institution: Kyoto University
• Principal Investigator: Asano Yasuhito 京都大学, 情報学研究科, 特定准教授 (20361157)
• Research Collaborator: Zhang Xinpeng ; Shimizu Toshiyuki
• Project Period (FY): 2015-04-01 – 2019-03-31
• Keywords: 知識発見 / 知識体系化 / エンティティ間の関係

Outline of Final Research Achievements

We have studied fundamental techniques for discovering and organizing knowledge about implicit relationships between concepts. We have also studied applications of connecting different kinds of data utilizing implicit relationships of concepts contained in the data. Implicit relationships between concepts are different from explicit relationships such as "A is a part of B" which are described explicitly on documents or data; they are relationships obtained by analyzing documents or data and connecting discovered explicit relationships found on several parts. We have established network analysis methods and text analysis techniques required as fundamental techniques for discovering such relationships. We have also established several analysis methods for particular kinds of data to apply the techniques above.

Free Research Field

情報学

Academic Significance and Societal Importance of the Research Achievements

(1) 本研究におけるネットワーク分析技術の構築の結果，潜在的関係の発見以外にも応用できる手法が開発できた．例えば，ライドシェアリング等で相乗りする際の費用を軽減するのに有効であると思われる，合流経路の最適化手法を開発できた．
(2) 様々な異種データを実際に結びつける手法を提案することができた．例えば，商品のレビューデータと商品の機能(性能)データから「商品の機能とユーザの使用目的」の関係(デジカメの例を挙げれば，鳥の撮影という目的に必要な機能や，夜景撮影という目的に必要な機能等がある)についての知識を発見する手法を提案できた．