Identifying and locating defects in software projects is a difficult task. Further, estimating the density of defects is more difficult. Measuring software in a continuous and disciplined manner brings many advantages such as accurate estimation of project costs and schedules, and improving product and process qualities. Detailed analysis of software metric data gives signijicant clues about the locations and magnitude of possible defects in a program. The aim of this research is to establish an improved method for predicting software quality via identifying the defect density of fault prone modules using machine-learning techniques. We constructed a two-step model that predicts defect density by taking module metric data into consideration. Our proposed model utilizes classification and regression type learning methods consecutively. The results of the experiments on public data sets show that the two-step model enhances the overall performance measures as compared to applying only regression methods.
|Title of host publication||EUROMICRO 2007 - Proceedings of the 33rd EUROMICRO Conference on Software Engineering and Advanced Applications, SEAA 2007|
|Number of pages||8|
|Publication status||Published - 1 Dec 2007|
|Event||33rd EUROMICRO Conference on Software Engineering and Advanced Applications, SEAA 2007 - Lubeck, Germany|
Duration: 27 Aug 2007 → 31 Aug 2007
|Conference||33rd EUROMICRO Conference on Software Engineering and Advanced Applications, SEAA 2007|
|Period||27/08/07 → 31/08/07|