Software Defect Prediction Analysis Using Machine Learning Techniques

There is always a desire for defect-free software in order to maintain quality customer satisfaction and save testing expenses. As result, we examined various known ML techniques optimized on freely available data set. The purpose of the research was improve model performance terms accuracy precision dataset compared previous research. investigations show, can be further improved. For this purpose, employed K-means clustering categorization class labels. Further, applied classification models selected features. Particle Swarm Optimization utilized optimize models. We evaluated through precision, accuracy, recall, f-measure, error metrics, confusion matrix. results indicate that all achieve maximum results; however, SVM outperformed with highest achieved 99% 99.80%, respectively. NB, Optimized RF, RF ensemble approaches are 93.90%, 93.80%, 98.70%, 99.50%, 98.80% 97.60, In way, studies, which our goal.