Accuracy Evaluation of C4.5 and Naïve Bayes Classifiers Using Attribute Ranking Method

Abstract

This paper intends to classify the Ljubljana Breast Cancer dataset using C4.5 Decision Tree and Naïve Bayes classifiers. In this work, classification is carriedout using two methods. In the first method, dataset is analysed using all the attributes in the dataset. In the second method, attributes are ranked using information gain ranking technique and only the high ranked attributes are used to build the classification model. We are evaluating the results of C4.5 Decision Tree and Naïve Bayes classifiers in terms of classifier accuracy for various folds of cross validation. Our results show that both the classifiers achieve good accuracy on the dataset.