An Efficient Method for Assessing the Impact of Refactoring Candidates on Maintainability Based on Matrix Computation

Abstract

For automating refactoring identification, previous methods for assessing the impact of a large number of refactoring candidates may be computationally expensive. In our paper, we propose an efficient method for assessing the impact of refactoring candidates on maintainability based on matrix computation, which is approximate but fast. This proposed method is evaluated on a refactoring identification approach for Edit and Columba, two large-scale open source projects. The experiments show that the proposed method requires less time for assessing refactoring candidates and that the refactoring identification approach using our proposed method also improves maintainability.