Identification of Web Service Refactoring Opportunities as a Multi-objective Problem
Hanzhang Wang, Ali Ouni, Marouane Kessentini, Bruce R. Maxim, William I. Grosky · 2016
We propose, in this paper, to consider the problemof Web service antipatterns detection as a multi-objectiveproblem where examples of Web service antipatterns and welldesignedcode are used to generate detection rules. To thisend, we use multi-objective genetic programming (MOGP)to find the best combination of metrics that maximizes thedetection of Web service antipattern examples and minimizesthe detection of well-designed Web service design examples. We report the results of an empirical study using 8 differenttypes of common Web service antipatterns. We compared ourmulti-objective formulation with random search, one existingmono-objective approach, and one state-of-the-art detectiontechnique not based on heuristic search. Statistical analysis ofthe obtained results demonstrates that our approach is efficientin antipattern detection, on average, with a precision score of94% and a recall score of 92%.