T-way Test Suite Generation Strategy based on Ant Colony Algorithm to Support T-way Variable Strength

Abstract

T-way test suite generation strategy based on Ant Colony algorithm (TTSGA) has been developed to support t-way variable strength testing which tackles exhaustive testing issues. It employs the ant colony optimization algorithm to generate near-optimal number of test suite size. Even though the test suite size is smaller than exhaustive testing, the strategy covers every possible combination of interacting parameters. The strategy has been evaluated by using benchmarked experiments. Results obtained were compared with other existing strategies that support variable strength. It was found that TTSGA produces comparable results with other existing strategies especially for higher strength configurations. Two non-parametric tests, which are Wilcoxon Rank and Friedman test, have been conducted to analyze the results statistically between TTSGA and HSS as only both strategies have complete experiments results. Although the results shows that there is no significant difference of test suite size among them, TTSGA is in the first rank in the Friedman test.

Ontology-based Negative Selection Algorithm For Pairwise Testing

“…N Ramli et al [2] proposed the metaheuristics search techniques for the Ant Colony Algorithm, which demonstrates it produces a different number of test cases depending on the test configuration. Using a Gravitational Search Algorithm (GSA), K. Htay et al [3] demonstrated that the test case produced is the best and that all covered tuples are removed from the list to prove that the test case produced is the best.…”