Intelligent Data Engineering and Automated Learning — IDEAL 2000. Data Mining, Financial Engineering, and Intelligent Agents

Abstract

Topic spotting is the task of assigning a category to the document, among the predefined categories. Topic spotting is called text categorization. Controlled indexing is the procedure of extracting the informative terms reflecting its contents, from the text. There are two kinds of repositories, in the proposed scheme of topic spotting; one is the integrated repository for controlled indexing and the other is topic repository for topic spotting. Repository is constructed by learning the texts, and consists of terms and their associated information: the total frequency and IDF (Inverted Document Frequency). An unknown text is represented into the list of informative terms by controlled indexing referring the integrated repository and the category corresponding to the largest weight is determined as the topic (category) of the text. In order to validate, the news articles from the site, ‘http://www.newspage.com” are used as examples, in the experiment of this paper.