Software Fault Prediction Using Two -Stage Data Preprocessing Method

Abstract

Software testing plays important role in software development life cycle (SDLC) while dealing with critical business applications. From SDLC 70% of time and cost is used for testing only, it is time consuming and costly phase. For this one of the cost effective technique can be used which predicts a module as fault-prone and non-fault prone prior to testing. Predicting a safe module during testing and after development increases the cost of projects. As need to allocate better-test resources for module. This process may take most of the time in preparation and may lose some modules untested which cause accidental failure. Defective software modules cause software failures, increase development and maintenance costs, and decrease customer satisfaction. In this research, we present a novel fault prediction technique that reduces the probability of false (pf) and increases the precision for detection of faulty modules. Software Reliability is becoming an essential attribute of any software system. To improve software quality and testing efficiency by constructing predictive models from code attributes to enable a timely identification of fault-prone modules.Software defect prediction model helps in early detection of faults and contribute to their efficient removal and producing a reliable software system.