WAP: A Novel Automatic Test Generation Technique Based on Moth Flame Optimization

In this work, we present a novel technique forautomatic test data generation that generates the whole test suitein a single run. The novelty of our proposed technique lies mainlyin (i) using Moth Flame Optimization (MFO) algorithm for thefirst time in automatic test generation, and (ii) introducing theuse of a generic objective function that is independent of themethod under test (MUT). The proposed objective functiondynamically evaluates the fitness of each solution, i.e. test case, with respect to the so-formed test suite, based on the effectivenessof adding this test case to the currently formed test suite. Thisdynamic approach replaces the use of a static fitness functionthat evaluates the fitness of each solution independently of othersolutions. The proposed technique tries to find small-sized testsuite with maximum coverage by iteratively eliminating test casesthat do not contribute to the overall coverage of the test suite. The results show that our technique is better than the randomgenerator with improvement up to two orders of magnitude. Italso outperforms Genetic Algorithm (GA) in four out of fivebenchmark methods achieving improvements between 74% and 83% in the number of generated test cases.