A Software Fault Prediction on Inter- and Intra-Release Prediction Scenarios

Abstract

Software quality engineering applied numerous techniques for assuring the quality of software, namely testing, verification, validation, fault tolerance, and fault prediction of the software. The machine learning techniques facilitate the identification of software modules as faulty or non-faulty. In most of the research, these approaches predict the fault-prone module in the same release of the software. The model is found to be more efficient and validated when training and tested data are taken from previous and subsequent releases of the software respectively. The contribution of this paper is to predict the faults in two scenarios (i.e., inter- and intra-release prediction). The comparison of both intra- and inter-release fault prediction by computing various performance matrices using machine learning methods shows that intra-release prediction has better accuracy compared to inter-releases prediction across all the releases. Also, both the scenarios achieve good results in comparison to existing research work.