A Study on Information Alignment by Composite Generative Model

2014 Fiscal Year Final Research Report

• Project/Area Number: 23300040
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Media informatics/Database
• Research Institution: National Institute of Informatics
• Principal Investigator: TAKASU Atsuhiro 国立情報学研究所, コンテンツ科学研究系, 教授 (90216648)
• Co-Investigator(Kenkyū-buntansha): MASADA Tomonari 長崎大学, 工学研究科, 准教授 (60413928)
• Co-Investigator(Renkei-kenkyūsha): FUKAGAWA Daiji 同志社大学, 文化情報学部, 助教 (10442518)
• Project Period (FY): 2011-04-01 – 2015-03-31
• Keywords: トピックモデル / 情報推薦 / 機械学習

Outline of Final Research Achievements

The purpose of this study is to develop topic models for analyzing information in various aspects. We first develop a topic model for handling time as well as text, where we add timestamps to each document. The model generates both text and timestamps simultaneously. Next we extend the model to treat networked documents where documents are linked each other like citations of academic papers. We apply the models to researcher recommendation systems and empirically show that features extracted by the models are effective for recommendation.

Free Research Field

情報工学