Multi-objective Test Case Selection Through Linkage Learning-Based Crossover

Test Case Selection (TCS) aims to select a subset of the test suite run for regression testing. The selection is typically based on past coverage and execution cost data. Researchers have successfully used multi-objective evolutionary algorithms (MOEAs), such as NSGA-II its variants, solve this problem. These MOEAs use traditional crossover operators create new candidate solutions through genetic recombination. Recent studies in numerical optimization shown that better recombinations can be made using machine learning, particular linkage learning. Inspired by these recent advances field, we propose variant NSGA-II, called L2-NSGA, uses learning optimize case selection. In particular, an unsupervised clustering algorithm infer promising patterns among (subset suites). Then, are next iterations L2-NSGA preserve inferred patterns. Our results show our customizations make more effective sub-sets generated less expensive detect faults than those literature