Support vector machine based fuzzy classification model for software fault prediction

Abstract

Defect proneness prediction of software modules always attracts the developers because it can reduce the testing cost as well as software development time. In the current context, with constantly increasing constraints like requirement ambiguity and complex development process, developing a fault free reliable software is a daunting task. To deliver reliable software, software engineers are required to execute exhaustive test cases which become tedious and costly for software enterprises. To ameliorate the testing process one can use a defect prediction model so that testers can focus their efforts on defect prone modules. Software defect prediction models use historical defect database to forecast error-prone modules. Defect prediction models require empirical validation to ensure their relevance to a software company. In this paper, a new Support Vector Machine based Fuzzy classification based prediction model has been proposed and evaluated on bug data base of an open source software project. In the proposed model a rule base is constructed using support vectors and the membership grade is calculated using Gaussian membership functions. Rule set optimization is done using Genetic algorithm. It is found that the proposed model gives very promising results on the criteria of probability of bug detection, probability of false alarm and accuracy.