Software Defect Prediction Using Software Metrics with Naïve Bayes and Rule Mining Association Methods

Abstract

Software defect prediction (SDP) can help testers decide allocation of resources rationally to find defects effectively, so as to improve software quality. Naive Bayes (NB) is one of the most used classification algorithms because of the simplicity of the algorithm and easy to implement. The purpose of this study is to add the process of selecting features using ARM in the software prediction process using the NB method in the hope that it can improve the performance of the method using software metrics. Software metrics have an association with one another in completing software, so this cannot be ignored. Results of the empirical evaluation of scenario 1 (one) showed an increase with the values of parameter precision, recall, f-measure and accuracy of 0.101, 0.190, 0.154 and 0.180, and scenario 2 (second) also increased by 0.106, 0.182, 0.159 and 0.163, also as in scenario 3 (third) the proposed method shows good performance compared to using SVM, NN and DTREE with an average performance of 0.960 while the others are 0.855, 0.859 and 0.861. From the empirical results of the three scenarios made, the proposed performance method is better than the other methods.