Search-Based Refactoring Using Recorded Code Changes

Over the past decades, many techniques and tools have been developed to record the sequence of applied refactoring to improve design quality. We start from the observation that these recorded code changes can be used to propose new refactoring solutions in similar contexts. In addition, this knowledge can be combined with structural and semantic information, used by existing work, to improve the automation of refactoring. In this paper, we propose a multi-objective optimization approach to find the best sequence of refactorings that maximizes the use of refactoring applied in the past to similar contexts, minimizes semantic errors and minimizes the number of defects (improve code quality). To this end, we use the non-dominated sorting genetic algorithm (NSGA-II) to find the best trade-off between these three objectives. We report the results of our experiments on different open source java projects.