Project Name

News Article Analysis Group

Member

Norifumi Hirata, Tomotaka Tsujino, Kenta Kato, Ryoji Suzuki, Michisato Koyama

Keyword

Topic Analysis，Natural Language Processing, Text Mining, NewsML

Purpose

Topic analysis for news articles on Web.

Outline

The notion of a "topic" was modified and sharpened to be an "event". An event means some unique thing that happens at some point in time.

It takes much computation time for topic analysis of many articles. But actually the topics which users need is few. To improve the computation time, we need to focus on users' preferences. We propose a system to analyze topics based on users' preference. The system select articles by users' preferences. And the system analyzes topics which are consisted by a few articles at a short time.