Predicting Upgrade Project Defects Based on Enhancement Requirements: An Empirical Study
In upgrade project development, Enhancement Requirements (ER, e.g. requirement additions and modifications) introduce new defects to the project. We need to evaluate this impact to help plan later project schedule and resources. Typically, many of the existing prediction technologies estimate defects based on software size or process performance baselines. However, they are limited in estimating the impact of ER on product quality. This paper proposes a novel ER-based defect prediction method using information retrieval (IR) technique and support vector machines (SVM). We analyze the historical data of defects and requirement specifications of actual upgrade projects to establish multiple prediction models to estimate new defects introduced by ER. Then we design two experiments to validate the method and report some preliminary results. The results indicate that our method can provide useful support for impact analysis of requirement evolution in upgrade projects.
KeywordsDefect Prediction Enhancement Requirement Information Retrieval Support Vector Machines
Unable to display preview. Download preview PDF.
- 4.Boehm, B.W., Horowitz, E., Madachy, R., et al.: Software Cost Estimation with COCOMO II. Prentice Hall PTR, Upper Saddle River (2000)Google Scholar
- 8.Malaiya, Y.K., Denton, J.: Requirements Volatility and Defect Density. In: 10th International Symposium on Software Reliability Engineering, p. 285 (1999)Google Scholar
- 9.Xing, F., Guo, P., Lyu, M.: A novel method for early software quality prediction based on support vector machine. In: Proc. of the 16th IEEE Int’l Symp. on Software Reliability Engineering (ISSRE 2005), pp. 213–222 (2005)Google Scholar
- 10.Gospodnetic, O., Hatcher, E.: Lucene in Action. Maning Publication (2006)Google Scholar
- 11.Qin, J., Lu, R.Z.: Feature Extraction in Text Categorization. Journal of Computer Applications 23(2), 45–46 (2003)Google Scholar
- 13.Gunn, S.R.: Support Vector Machines for Classification and Regression, Technical Report. Faculty of Engineering, Science and Mathematics School of Electronics and Computer Science, University of Southampton (1998)Google Scholar