Abstract

Achieving bug-free software is hard as most of the time there are hidden defects. Predicting the rate of faultiness in software modules before commercializing is still a challenge to software engineers. Early detection of fault prone modules not only helps the software project manager to get good quality software, but also to sustain their customer satisfaction. In this paper, two classifiers and a meta-heuristic feature selection based models are proposed for software fault prediction. We perform a comparison study of different classifiers such as Support Vector Machine, Random Forest and meta-heuristic feature selection technique such as Particle Swarm Optimization (PSO). We have conducted experiments on nine data sets collected from GitHub Repository. In this study F-measure, accuracy and AUC are considered as performance measures. We found that, the classification accuracy of proposed model is not degraded, after reducing 75.15%, of the total features on an average, by PSO. We also found that the prediction performance of Random Forest is better than other model such as Support Vector Machine.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call