期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2014
卷号：5
期号：2
页码：2348-2354
出版社：TechScience Publications
摘要：Software testing plays a vital role in software development especially when the software developed is mission, safety and business critical applications. Software testing is the most time consuming and costly phase. Prediction of a modules info fault-prone and non fault prone prior to testing is one of the cost effective technique. Predicting a safe module as faulty increases the cost of projects by more cautious and better-test resources allocation for those modules, whereas prediction of faulty code as fault free code end up in under-preparation and may leave modules untested this may cause accidental failure and lead towards massive loss . In this research, we present a novel fault prediction technique that reduces the probability of false alarm (pf) and increases the precision for detection of faulty modules. The general expectation from a predictor is to get very high probability of false alarm (pf) to get more reliable and quality software product. We have taken embedded systems software for this study and the goal is to predict as many faulty modules as possible. In this paper we apply a supervised discretization for pre-processing and clustering based classification for prediction of a modules info fault-prone (fp) and non faultprone (nfp) modules. To evaluate this approach we perform an extensive comparative experimental study for the effectiveness of our method with benchmark results for the same embedded software’s. Our fault prediction model produces better results than the standard and benchmark approaches for software fault prediction. Results from our proposed model significantly decreases probability of false alarm (pf) down to 9% while increasing precision and balance rates at 68% and 79% respectively
