Software Defect Prediction across Multiple Datasets using Classification Techniques

Abstract

             Defect prediction in the early phases of the software development life cycle is a critical activity of the quality assurance process that has been extensively researched over the last two decades. Early detection of faulty modules in software development can assist the development team in making efficient and effective use of available resources to provide high-quality software products in a short period of time. The machine learning technique, which works by detecting hidden patterns among software features, is an excellent way to discover problematic modules. The results showed that Priority Based Fuzzy SVM yields 96% of accuracy while different defect datasets are taken into account compared to other techniques such as decision tree, PART, Random Forest, Naive Bayes and SVM. The mentioned existing techniques are used for prediction using same data sets which are taken for proposed work.