A Study on Classification of Documents using Correspondence Analysis

Abstract

The opportunity that we search target document(s) among a large amount of electronic documents such as WEB information and those by document scanner will be more increasing in near future. We focus on the approach of the classification of documents based on the similarity of the distribution pattern of the keyword frequency among them. In this paper, we propose the method that uses the correspondence analysis for the classification of documents by the similarity of the keyword pattern. We apply the proposed method to the patents of Nagoya University, and report the result in terms of the accuracy of the classification.