Various methods for software fault prediction

Fault estimation of a software module is the probability that the module contain the error and it is a defect which will cause the further failures in the software so the early detection of these errors will help the experts or developers to concentrate on these errors which will improve the quality of software also. As the demand for producing the software quality increases day-by-day, so early estimation of errors in any module is a great step towards it. It will predict the quality attributes like reliability, maintenance, efforts for testing, wrong syntax, wrong logic, misleading documentation, productivity and testing efforts. As the demand for quality of the software increased so the development of the machine learning methods to explore the data are also increased. Early prediction of the faults will make the researchers or experts to give there best in the faulty areas or they can save there much time which is going to be waste in finding the fault prone area in software. These faulty area or modules in the software can cause the failure in extended version. So in this study errors are estimated by using LCOM, BPA, ROC, PCM, DBSCAN.