Combinatorial test case prioritization using hybrid Energy Valley Dwarf Mongoose Optimization approach

Combinatorial Test Case Prioritization is a technique used in software testing to improve the efficiency and effectiveness of test suites. It involves selecting and ordering test cases based on their ability to detect faults, especially those caused by interactions between multiple parameters. Artif...

Full description

Saved in:

Bibliographic Details
Published in:	Expert systems with applications Vol. 271; p. 126634
Main Authors:	Kanagaraj, Kamaraj, Nithiyanandam, Prasath, Sekar, Saradha, Shanmugam, Sangeetha
Format:	Journal Article
Language:	English
Published:	Elsevier Ltd 01.05.2025
Subjects:	Deep Q Net (DQN) Dwarf Mongoose Optimization Algorithm (DMOA) Energy Valley Optimization (EVO) Regression Testing Test Case Prioritization (TCP) Dwarf Mongoose Optimization Algorithm (DMOA) Regression Testing Energy Valley Optimization (EVO) Test Case Prioritization (TCP) Deep Q Net (DQN)
ISSN:	0957-4174
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Combinatorial Test Case Prioritization is a technique used in software testing to improve the efficiency and effectiveness of test suites. It involves selecting and ordering test cases based on their ability to detect faults, especially those caused by interactions between multiple parameters. Artificial Intelligence (AI) has made significant contributions to Combinatorial Test Case Prioritization (TCP) by introducing advanced techniques to enhance the efficiency of the testing process. However, managing dependencies between test cases and adjusting the prioritization accordingly can be complex and time-consuming in most of the previous techniques. Therefore, an Energy Valley Dwarf Mongoose Optimization Algorithm (EVDMOA) is devised for Combinatorial TCP. Initially, the software programs are collected from the dataset. Then, test case generation is performed to create the test suites. Next, the combinatorial TCP is performed. Here, the fitness parameters such as Average Percentage of Fault Detected (APFD), Average Percentage of Branch Coverage (APBC), and weight are considered for fitness evaluation. Moreover, the weights in the fitness function are computed by the Deep Q Net (DQN), which is trained by the proposed EVDMOA. At last, the prioritized test cases are obtained. The EVDMOA achieves the AFPD, APBC, and fitness values of 0.907, 0.914, and 0.926. Moreover, the EVDMOA helps in maintaining the overall quality and reliability of the software.
ISSN:	0957-4174
DOI:	10.1016/j.eswa.2025.126634