Systematic Review of Bespoke Techniques of Software Fault Prediction: Machine Learning and Conventional Techniques
Main Article Content
Abstract
Software fault prediction is the practice of creating frameworks that software professionals are able to employ in the early stages of the software development life cycle to detect troubling constructs related to modules or classes. Several techniques have been suggested in the past for this purpose, the majority of which relied on lengthy mathematical models that did not appear to be suited for the current challenge. With technological advances, artificial intelligence (AI) modeling, also known as soft computing approaches, unlike conventional models, is gaining attention in research as it requires less processing and anticipates results quickly. In this paper, the author attempts to review conventional and machine learning techniques used previously for software fault prediction. For this review, data has been collected from various reputed library, such as IEEE Xplore. The present article has analyzed the various papers published from 2018 onward on the use of various conventional and non-conventional techniques. The extensive study’s analysis given in this article will be of great interest to academicians and professionals working in the field of software fault prediction in determining in which situations they need to apply which technique for the best results.