Predicting Bug in a Software using ANN Based Machine Learning Techniques

To achieve a successful software, bug prediction mainly place an important role. Nowadays, bug prediction has become an essential argument in the maintenance and development phase. So, it's necessary to predict bugs in earlier stages of software development life cycle. The challenge here is to develop a model that helps in predicting bugs leading to good quality, reliable, efficient and cost-effective software. For complex software projects, bugs are major issues. The proposed model provides the comparative analysis on various machine learning algorithms that are developed to predict bugs namely Random Forest, Logistic regression, Decision Tree, Artificial Neural Network and Naïve Bayes. Here Artificial Neural Network is used along with other algorithms such that the model can be trained for large datasets, in order to get more accurate results working effectively for various scenarios. The performance of each model is evaluated, and cross validation is performed followed by visualizing the results. Finally, when all the models are compared Artificial Neural Network appears to be the best model by providing 82.77%.