Software Maintenance Prediction Using Regression Models

Software Maintenance is of utmost importance for any industry. So, to predict the value of software maintenance beforehand also becomes very important, hence many software maintenance prediction algorithms has been devised previously. The software maintenance prediction values by taking current and previous datasets of the same software and comparing the lines of code in both of them. In this paper, performance of various machine learning algorithms and ensemble learning using deep neural networks has been com-pared. The machine learning algorithms used here are Elastic Net Regression, Gaussian Process Regression, Lasso Regression, Ridge Regression, Least Angle Regression. Various performance metrics used are RMSE, MAE and MSE. By observing the perofemance of the regression models, Elastic Net Regression and Lasso Regression are perofrmed better tahn otehr models. The Elastic Net and Lasso Regression models achieved RSMSE values are 400.87, 146.86, 80.17,168.79, 272.46 withrespetive datsets.