Abstract

Rapid development of software technology has influence on substantial industrial growth. Wide application of software in business related matters leads to development of reliable and defect free software system which is a challenging task. It requires development of effective techniques for prediction of software defects at early stage. For complexities in manual prediction of defects, automated techniques have come into effect. They are basically based on learning of pattern from earlier versions of software development and finding out the defects from the current version. Considerable impact of these techniques on industrial growth by predicting defects in software system attracted researchers in this field.In-spite of many studies performed by applying these techniques, desirable performance level and accurate defect prediction still remains a challenging task. For solving this problem, a hybrid technique based on Nonlinear Manifold Detection Techniques (Nonlinear MDTs) and machine learning for prediction of defects has been proposed in this paper. A new hybrid Nonlinear Manifold Detection (Nonlinear MD) Model has been applied for selecting and optimizing the features of software datasets that have been processed using Decision Tree (DT) and Random Forest (RF) classifications. Finally, a comparison and statistical evaluation of the experimental results obtained using new hybrid Nonlinear MD Model-DT have been made by Friedman test followed by Wilcoxon Sign rank test. The statistical outcome revealed that the proposed new hybrid Nonlinear MD Model-DT classification is better result oriented and more accurate in software defect prediction.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.