OPTIMIZED SOFTWARE FLAW FORECASTING SCHEME BASED ON DATA MINING AND DIFFERENTIAL EVOLUTION

Software flaws are affecting the software development process and performance which yields an unpredicted behavior and produce erroneous results. Software defects are expensive in terms of quality and cost. Detection and correction of these defects are significant task to ensure the quality of the software product. Numerous software repositories hold source code of large projects as many modules and hold data for the software metrics of these modules and the defective state of each module. In this paper, an optimized software flaw forecasting scheme is proposed based on Data Mining and Differential Evolution(DE). The data mining approach is used to select the attributes that forecast the defective state of software modules. The DE is utilized for optimization process. The experimental results are presented to exhibit the better prediction competence of the proposed scheme.