Effective Feature Selection for Preprocessing Step of Classification Using Modified-MCA

Abstract

A novel metric that integrates the correlation and reliability information between each feature and each class obtained from Multiple Correspondence Analysis (MCA) is currently the popular solution to score the features for feature selection. However, it has the disadvantage that p-value which examines the reliability is conventional confidence interval. The main goal of this paper is to introduce a new classifier independent (filter-based) feature selection method, Modified Multiple Correspondence Analysis (Modified-MCA) which is designed to modify MCA, improving the reliability. The efficiency and effectiveness of proposed method is demonstrated through extensive comparisons with MCA and other feature selection methods, using five benchmark datasets provided by WEKA and UCI repository. Naïve Bayes, Decision Tree and JRip are used as the classifiers. The classification results, in terms of classification accuracy and size of feature subspace, show that the proposed ModifiedMCA outperforms three other feature selection methods, MCA, Information Gain, and Relief