Effective and efficient software testing utilizes the minimum resources of1software. Therefore, it is important to construct the procedure which is not1only able to perform the efficient testing but also minimizes the utilization1of project resources. The goal of software testing is to find maximum1defects in the software system. As world is continuously moving toward1data driven approach for making important decision. Therefore, in this1research paper we performed the machine learning analysis on the publicly1available datasets and tried to achieve the maximum accuracy. The major1focus of the paper is to apply different machine learning techniques on the1datasets and find out which technique produce efficient result. Particularly,1we proposed an ensemble learning models and perform comparative1analysis among KNN, Decision tree, SVM and Na¨ive Bayes on different1datasets and it is demonstrated that performance of Ensemble method is1more than other methods in term of accuracy, precision, recall and F1-1score. The classification accuracy of ensemble model trained on CM1 is198.56%, classification accuracy of ensemble model trained on KM2 is198.18% similarly, the classification accuracy of ensemble learning model1trained on PC1 is 99.27%.