Nature-inspired metaheuristic methods in software testing
Niloofar Khoshniat,Amirhossein Jamarani,Ahmad Ahmadzadeh,Mostafa Haghi Kashani,Ebrahim Mahdipour
DOI: https://doi.org/10.1007/s00500-023-08382-8
IF: 3.732
2023-06-08
Soft Computing
Abstract:Software quality is becoming a momentous challenge in software engineering processes, and software testing has a pivotal role in its measurements. Nature-inspired metaheuristic methods play an essential role in software testing, in which various studies have been conducted in this field; however, due to a lack of wide-ranging papers reviewing these methods, conducting a comprehensive systematic review to examine an array of crucial mechanisms in this field has become a necessity. This study aims to present a detailed analysis and taxonomically classifies the metaheuristic methods inspired by nature. This paper compromises a systematic literature review of 65 chosen studies published between 2015 and 2022. Genetic algorithm-based, hybrid, ant colony optimization-based, cuckoo search-based, firefly algorithm-based, artificial bee colony-based, and other metaheuristic methods constitute this systematic study’s stratification. Evaluation methods, applied tools, merits, and demerits of each reviewed article are investigated. Additionally, future directions and open issues are addressed. This conducted paper not only expounds on software testing strengths, open issues, and future works, but also recognizes the quest for optimizing the insufficient metrics in software testing, such as mutation score, complexity, and scalability, which would be the propulsion of the testing process if consummated.
computer science, artificial intelligence, interdisciplinary applications