Exploring Software Quality Classification with a Wrapper-Based Feature Ranking Technique

Feature selection is a process of selecting a subset of rel-evant features for building learning models. It is an impor-tant activity for data preprocessing used in software quality modeling and other data mining problems. Feature selec-tion algorithms can be divided into two categories, feature ranking and feature subset selection. Feature ranking or-ders the features by a criterion and a user selects some of the features that are appropriate for a given scenario. Fea-ture subset selection techniques search the space of possi-ble feature subsets and evaluate the suitability of each. This paper investigates performance metric based feature rank-ing techniques by using the multilayer perceptron (MLP) learner with nine different performance metrics. The nine performance metrics include Overall Accuracy (OA), De-fault F-Measure (DFM), Default Geometric Mean (DGM), Default Arithmetic Mean (DAM), Area Under ROC (AUC)