Ensemble Classifiers in Software Defect Prediction: A Systematic Literature Review

The Software Defect Prediction area is important in software development since it allows anticipating possible software defects before they occur in production and affect end users. If software defects are not corrected, they can degrade system quality and reliability. This work aims to carry out a Systematic Literature Review to know which ensemble classification algorithms help predict if a system module is prone to software defects. In the same way, it is intended to know the most popular metrics that evaluate the effectiveness of the applied algorithms and the results of the algorithms in the area of Software Defect Prediction. This work shows that the most used ensemble classifiers for predicting software defects are Random Forest and AdaBoost; regarding the metrics, most of the studies use AVC and F-Measure and Accuracy.