Predicting fault prone modules by the Dempster-Shafer belief networks

• Published in: 18th IEEE International Conference on Automated Software Engineering, 2003. Proceedings Vol. 2003; pp. 249 - 252
• Main Authors: Guo, Lan, Cukic, Bojan, Singh, Harshinder
• Format: Conference Proceeding Journal Article
• Language: English
• Published: Piscataway, NJ, USA IEEE Press 01.10.2003; IEEE
• Series: ACM Conferences
• Subjects: Accuracy; Classification tree analysis; Computing methodologies; Computing methodologies > Machine learning; Fault diagnosis; Lab-on-a-chip; Logistics; NASA; Power system modeling; Predictive models; Software quality; Statistics
• DOI: 10.1109/ASE.2003.1240314
• ISBN: 0769520359; 9780769520353
• ISSN: 1938-4300; 1527-1366

This paper describes a novel methodology for predicting fault prone modules. The methodology is based on Dempster-Shafer (D-S) belief networks. Our approach consists of three steps: First, building the Dempster-Shafer network by the induction algorithm; Second, selecting the predictors (attributes) by the logistic procedure; Third, feeding the predictors describing the modules of the current project into the inducted Dempster-Shafer network and identifying fault prone modules. We applied this methodology to a NASA dataset. The prediction accuracy of our methodology is higher than that achieved by logistic regression or discriminant analysis on the same dataset.