Research Article
Simultaneous Localization of Multiple Defects in Software Testing Based on Reinforcement Learning
@INPROCEEDINGS{10.1007/978-3-030-82562-1_16, author={Jiajuan Fang and Yanjing Lu}, title={Simultaneous Localization of Multiple Defects in Software Testing Based on Reinforcement Learning}, proceedings={Multimedia Technology and Enhanced Learning. Third EAI International Conference, ICMTEL 2021, Virtual Event, April 8--9, 2021, Proceedings, Part I}, proceedings_a={ICMTEL}, year={2021}, month={7}, keywords={Reinforcement learning Software testing Multiple defects Positioning Genetic algorithm}, doi={10.1007/978-3-030-82562-1_16} }- Jiajuan Fang
Yanjing Lu
Year: 2021
Simultaneous Localization of Multiple Defects in Software Testing Based on Reinforcement Learning
ICMTEL
Springer
DOI: 10.1007/978-3-030-82562-1_16
Abstract
At present, most software defect localization methods focus on single defect localization, but few on multi-defect localization. Therefore, the multi-defect localization method based on reinforcement learning is proposed. By using genetic algorithm, the candidate distribution population can be transformed into a sort of suspicious value of real program entity, and the location of multiple defects in software testing can be realized simultaneously. Experimental results show that, compared with the average evaluation index of the existing methods, the evaluation index(EXAM{F})of the proposed method is reduced by 1.19 and(EXAM{L})reduced by 1.05, which shows that the proposed method has better positioning performance and is suitable for popularization.
