Bayesian Network Based Bug-fix Effort Prediction Model
Development and use of prediction model (process performance models, PPM) are the primary requirements of high maturity practices. PPMs are useful tools for project management and process management. They help project managers to predict process performance with a known level of confidence thereby enabling them identify the risk and take actions. Over a last few years, Bayesian Belief Networks (BBN) have received a great deal of attention as prediction models, since they provide better solution to some of the problems found in Software Engineering when compared with traditional statistical models. In this paper, we are presenting our experience of using BBN for bug-fix effort prediction.
KeywordsProcess performance model BBN effort prediction
Unable to display preview. Download preview PDF.
- 1.CMMI Product Team, CMMI® for Development, Version 1.3, Software Engineering Institute (2010)Google Scholar
- 2.Automotive SIG, Automotive SPICE© Process Assessment Model (v2.5) and Process Reference Model V4.5 (2010) Google Scholar
- 4.Heckerman, D., Mamdani, A., Wellman, M.P.: Real-World Applications of Bayesian networks. Communications of the ACM 38(3) (March 1995)Google Scholar
- 5.Rodriguez, D., Harrison, R., Satpathy, M.: An Investigation of Prediction Models for Project Management. In: IEEE COMPSAC 2002 (2002)Google Scholar
- 6.BBN online Tutorial, http://www.eecs.qmul.ac.uk/~norman/BBNs/BBNs.html
- 7.Sarkar, S., Sindhgatta, R., Pooloth, K.: A Collaborative Platform for Application Knowledge Management in Software Maintenance Projects. In: Compute 2008, Bangalore, Karnataka, India, January 18-20. ACM (2008) ISBN 978-1-59593-950-0/08/01Google Scholar
- 8.Spirtes, P., Glymour, C., Scheines, R.: Causation, Prediction, and Search. Lecture Notes in Statistics, vol. 81. Springer, New York (1993) ISBN 0-262-19440-6PGoogle Scholar