A hybrid approach to improve the quality of software fault prediction using Naïve Bayes and k-NN classification algorithm with ensemble method

Abstract

This paper considers an improvisation in software fault prediction research area using supervised classification algorithms and it mainly focuses to increase the performance of fault prediction. In this paper, we propose a hybrid prediction model using Naive Bayes and k-nearest neighbour classification algorithm with vote ensemble method; in short it called as hNK. The goal of this model is to predict the best classification algorithm for software fault prediction based on the metrics and attributes of datasets. In the work, we have applied training sets and testing sets in hNK model with ensemble vote and we proposed the model to identify a suitable classification algorithm for fault prediction based on the accuracy and precision. We have achieved better results using hNK model for classifying supervised algorithms with different dataset.