Fuzzy Rules for Document Classification to Improve Information Retrieval

Abstract

In this work, we present a method to generate, from text documents, fuzzy rules used to classify documents and to improve the information retrieval. With this method, we face the issue of dimensionality in text documents for information retrieval. We also present a comparison analysis among the method that we proposed and well-known machine learning methods for classification. The aim of our work is to develop a mechanism to reduce the high dimensionality of the attribute-value matrix obtained from the documents and, consequently, scale up the proposed classifier. Some experiments have been run using different domains in order to validate the proposed approach and compare the results with the ones obtained with the OneR, K-Nearest Neighbor classifier, C4.5, Multi-variable Naive Bayes, and SVM methods. The experiments and the obtained results showed that this is a promising approach to deal with the dimensionality problem of document for information retrieval.