Iterative Feature Selection Using Information Gain & Naïve Bayes for Document Classification

Data Mining is the technique of analyzing large amount of data to determine the relation among large dataset. In this paper, we are discussing a new method for document classification. Usually Document classification has been done by using classifier algorithms. Naïve Bayes classifier is frequently used for classification which provides more accurate result for larger dataset. The usage of information gain with naïve Bayes classifier reduces the length of branches by selecting maximum gain which produces more accurate result in classification. We propose a new methodology of assigning weights using information gain with naïve Bayes classifier. The performance of naive Bayes learning with weighted gain increases accuracy than any other traditional methods using naïve Bayes. The experimental result indicates that the proposed system could improve the performance of naïve Bayes significantly.