Detection of Annotation Words and Construction of Structured Infomations based on Cohesiveness

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00429
• 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: University of Shizuoka
• Principal Investigator: Seiya OKUBO 静岡県立大学, 経営情報学部, 講師 (90422576)
• Co-Investigator(Kenkyū-buntansha): 池田 哲夫 静岡県立大学, 経営情報学部, 教授 (60363727) ; 風間 一洋 和歌山大学, システム工学部, 教授 (60647204) ; 斉藤 和巳 静岡県立大学, 経営情報学部, 教授 (80379544) ; 湯瀬 裕昭 静岡県立大学, 経営情報学部, 教授 (30240162)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 凝縮性 / 情報構造化 / アノテーションワード / 文書クラスタリング

Outline of Final Research Achievements

The purpose of this study is to classify document data and give annotation words for each classification. In order to achieve this purpose, we introduce an index called "Cohesiveness" that developed the concept of cluster coefficients, and proposed a classification and annotation word assignment method based on it. In order to evaluate the validity of the proposed method, we performed computer experiments and empirical evaluations. In the empirical evaluation, we applied it to the data set of KindaiShinbun articles. In computer experiments, validity was evaluated by applying it to various data existing on the Web. In addition, the scope of application was extended to data other than document data, and it was also applied to agricultural environmental data etc.From the above, the usefulness of the proposed method became clear.

Free Research Field

量子計算