Mutation based test generation using search based social group optimization approach

Mutation based test generation is a popular and effective process for creating the test suite that is appraised for its caliber over a pool of artificial faults. These artificial faults can be infused by imposing mutagenic rules that further assist meta-heuristic techniques for searching the evolved test suite in search space. Meta-heuristic techniques switch between multiple solutions in search space and result in an optimized solution. This paper implements and presents a new test set generation algorithm, SGO-MT, by embracing a recently developed search based approach, Social Group Optimization algorithm (SGO) for exposing numerous artificial faults in the software. It works on the principle of human learning nature from society and a teacher in the group. The efficacy of the proposed approach is measured on thirteen Java programs widely used in academia. The results demonstrate the good performance for finding the simple and stubborn faults.