Abstract: In software engineering, the most frequent problem highlighted by IT Practioners concerned the measurement of quality. In order to improve the quality of the software, fault prediction is the necessary task. This prediction reduces the time complexity between modules. In the recent years lot of software metrics are used for predicting whether the particular models of the software faulty are fault free. In this paper we have proposed K-Jensen Shannon Entropy Model based Clustering Algorithm for predicting the faults in software projects. In our experiment, we used CM1, PC1, KC1, KC2 and PC4 collected from NASA MDP. Finally, our proposed system is compared with Euclidean distance based K-Means Clustering Algorithm.
Keywords: software fault prediction, clustering, Quality and Metrics.