Software Defect Prediction using Cutting Edge Clustering Module and Machine Learning

The defect in the software arrives from building it and it can be solved by the default prediction which aims to give the outputs as the defection prone components or the code parts in which there is a defect during the testing stage of the development procedure. The main outcome of the prediction of the particular models is to see that testing is performed and all the defects that are present in the particular part of the software too comes into play. Traditionally the defect prediction focuses on the design features which are done manually and involves the complexity metrics to provide the input to the ML classifiers for finding the defects in a certain area or entire program. Instead, as per the trend, the features which are semantically and structural to the programs generally fail during the testing phase. Further, it should be detected properly to develop a proper software. This research work intends to showcase the cutting edge approach of the clustering module as a part of the machine learning technique and find the defects in the software.