Su Jeong Ko, Jung Hyun Lee

The KIPS Transactions:PartB , Vol. 8, No. 3, pp. 251-260, Jun. 2001
10.3745/KIPSTB.2001.8.3.251,   PDF Download:

Abstract

It is a problem that established Bayesian document categorization reflects the semantic relation inaccurately at feature expression of document. For the purpose of solving this problem, we propose Bayesian document categorizing method using Apriori-Genetic algorithm in this paper. Apriori algorithm extracts the feature of document being reflected semantics between words and constructs association word knowledge base through extracted association words. When association word knowledge base is constructed by Apriori algorithm, there are unsuitable association words in association word knowledge base. According to, it has a shortcoming that the accuracy of document categorization becomes lower. In order to complement a shortcoming, we use Genetic algorithm, which optimizes the association word knowledge base. Then, classifier using Bayesian probability categorizes documents based on optimized association word knowledge base. In order to evaluate performance of Bayesian document categorizing method using Apriroi-Genetic, we compare with Bayesian document categorizing method using Apriori algorithm and Bayesian document categorizing method using TFIDF and simple Bayesian document categorizing method.

Statistics
Cite this article