Prioritizing software regression testing using reinforcement learning and hidden Markov model

Abstract

AbstractSoftware regression testing is an essential testing practice that ensures that changes made to the source code of an application do not affect its functionality and quality. Within this research, we introduce a novel method for prioritizing software test cases using a fusion of reinforcement learning and hidden Markov model to enhance the efficiency of the testing process. The primary objective of this research paper is to maximize the likelihood of selecting test cases that have the highest priority of uncovering defects in new code changes introduced into the codebase. To assess the efficacy of our suggested methodology, we experimented on the test cases of five web applications. Our results demonstrate that our proposed approach can accurately identify critical test cases while minimizing false positives, as evidenced by an F1 score of 0.849. This outcome can help prioritize testing efforts, saving time, and resources while improving the overall efficiency of the testing process.Keywords: Regression testingtest case prioritization (TCP)hidden Markov model (HMM)reinforcement learning (RL) Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationNotes on contributorsNeelam RawatMs. Neelam Rawat is a dedicated research scholar in the field of Computer Science & Engineering at Sangam University. With an extensive portfolio that includes over 15 publications, 3 patents, and 2 authored books, she is actively engaged in pioneering research. Her primary areas of expertise lie in the domains of machine learning, deep learning, software testing, software engineering, quality assurance, and management.Vikas SomaniDr. Vikas Somani (PhD, M.Tech, MCA,BCA) has more than 16 years of Teaching and Industrial Experience. Currently he is Associate Professor and Assistant Dean, School of Engineering and Technology at the Sangam University, Bhilwara. He has diversified research interests in the areas of Cloud Computing, Artificial Intelligence, Machine Learning, Block chain and Internet of Things (IoT). He is a Member of IEEE, CSI, IAENG, ACM, IRED. He has published over 35 Research Paper in International, National Journal and Conferences and attended around 50 Workshops and STP. He has also Supervised/Guided more than 20 Research Work. Currently, under his 6 research scholars are working. He has Three Patent awarded and granted/design one from Government of India Patent Office and another from Germany Patent Office. He has also published Five Patents.Arun Kr. TripathiDr. Arun Kr. Tripathi has more than 21 years of Teaching experience and completed Ph.D. in Computer Applications with specialization in Wireless Networks. Presently he is appointed as Head of Computer Applications with and an additional responsibility of Head Cyber Security and Forensic Science Division. His major research interests are Computer Network, Network Security, IoT, Machine Learning etc. with over 70 published works in reputed Journals and Conferences. He reviewed more than 35 SCI-Indexed journal articles.