Hybrid regression testing technique: based on requirement priorities, fault and modification history

摘要

Context: The main challenge in incremental delivery of software is the testing of the increment with minimal number of test cases in such a manner that effect of execution of the subset is equivalent to the execution of whole test suite (regression testing). The execution of test suite subset must guarantee the capability of revealing all faults at the earliest. Objective: This paper aims to perform regression testing with minimal number of test cases that are capable of locating large percentage of faults. Method: The hybrid regression testing algorithm proposed in this paper is based on the clustering by employing three parameters: a) requirement priorities; b) fault detection history; c) module modification history. Results: The proposed technique was applied on live project of "Tool for automatic analysis and comparison of different release planning methods". The total of 14 requirements are considered for evaluation of proposed testing technique (three were added after the first review, leading to the process of reprioritisation). Conclusion: The evaluation of technique on real software revealed that the proposal resulted in better faults localisation with minimal number of test cases although its evaluation on a mature software system is kept as future work.