Extensive Software Fault Prediction: An Ensemble based comparison

In recent studies, fault prediction is used to envisage the software's quality via conventional machine learning techniques. There are numerous problems that distress the efficiency of the software like redundant data, irrelevance features and missing data. It is observed from the literature that performance of the machine learning methods varies from data-set to data-set. In this paper, authors examines the performance of the model get affected after applying several data processing techniques on different machine learning classifiers. Various performance measures like accuracy, RMSE and ROC (AUC) are considered for predicting the quality of software.