Author : Dharmendra Lal Gupta and Kavita Saxena

In this paper object oriented defect datasets have been collected from open source promise data repository. Out of 20 provided metrics in each dataset, most prominent 14 metrics have been selected using feature selection process. These metrics are directly responsible for bug prediction in such systems. In this paper most prompting classifier Logistic Regression based on 10-cross validation has been used. The findings have been analyzed using Area under Curve (AUC) values. This information can be used by software developers to enhance the quality of a system. WEKA tool has been used for finding and analysis of our result. A Comparative study has also been done in terms of AUC values obtained by proposed model and Mamdouh Alenezi Model. Our proposed model is providing better result that Mamdouh Alenezi Model.

Keywords: Software Bug, Software Defect Prediction, AUC

Article published in International Journal of Current Engineering and Technology, Vol.6, No.5 (Oct-2016)