查看文章

Techniques based on Machine Learning (ML) and various Artificial Intelligence (AI) algorithms are presently effectively deployed when dealing with real-world situations such as commercial business, digital service architectures, and manufacturing industries, though they are sparsely leveraged in the field of Software Testing. Because of the intricacy of this field, much AI/ML work on it is still academic. This article dispenses a swift summary of software testing's current status of development using machine learning and AI methods. The test types have been divided into groups of algorithms used in software testing. The study also seeks to establish links between the various AI methodologies and the types of tests to which they are applied. The emphasis of this novel study is on regression grey-box fuzzing which has substantial advancement implications when heuristics are considered. Hence by understanding the literature, the paper studies four different methodologies, namely QGA, FGDACO, GR-PSO, and QPSO.