Empirical comparison of regression methods for variability-aware performance prediction

Product line engineering derives product variants by selecting features. Understanding the correlation between feature selection and performance is important for stakeholders to acquire a desirable product variant. We infer such a correlation using four regression methods based on small samples of measured configurations, without additional effort to detect feature interactions. We conduct experiments on six realworld case studies to evaluate the prediction accuracy of the regression methods. A key finding in our empirical study is that one regression method, called Bagging, is identified as the best to make accurate and robust predictions for the studied systems.