A RULE-BASED PREDICTION METHOD FOR DEFECT DETECTION IN SOFTWARE SYSTEM

摘要

Software is a complex object that consists of different modules with changing degrees of defect occurrence. By efficiently and appropriate predicting the frequency of defects in software, software project managers can better utilize their workforce, cost and time to obtain better quality assurance. This paper proposes a rule-based prediction (RBP) method for defect detection and for planing the better maintenance strategy, which can support in the forecast a defective or non-defective software module before it can deploy for any software project. The RBP extends the Ripple-down rule (RDR) classifier method to construct an effective rule-basedmodel for accurately classifying the software defects. The method will enhance the software defect prediction so that software testers can spend more time in testing those components which are expected to contain errors. The experiment evaluation is performed over a software repository datasets and the obtained results showa satisfactory improvement.