Scaling up the Naive Bayesian Classifier using Genetic and Decision Tree for feature selection

Abstract

This paper aims to scale up the Naïve Bayesian Classifier using Genetic and Decision Tree for feature selection. The main reason is to predict patient's breast cancer result based on their diagnosis using this scaled classifier. Naïve Bayes can suffer from oversensitivity to redundant and/or irrelevant attributes. Several researchers have emphasized on the issue of redundant attributes, as well as advantages of feature selection for the Naïve Bayesian Classifier. In this paper, Genetic algorithm is used to reduce redundant attributes in feature selection, and then apply Decision tree algorithm to find an optimal set of feature weights that improve classification accuracy. By combining genetic algorithm with decision tree, and this method enhance the Bayesian classification to eliminate unnecessary features and produce fast, accurate classifiers. Bayesian classifier represents each class with a probabilistic summary, and finds the most likely class for each example it is asked to classify.