Prediction of Software Design Defect using Enhanced Machine Learning Techniques

Abstract

Prediction of software detection is most widely used in many software projects and this will improve the software quality, reducing the cost of the software project. It is very important for the developers to check every package and code files within the project. There are two classifiers that are present in the Software Package Defect (SPD) prediction that can be divided as Defect–prone and not-defect-prone modules. In this paper, the merging of Cost-Sensitive Variance Score (CSVS), Cost-Sensitive craniologist Score (CSLS) and Cost-Sensitive Constraint Score (CSCS). The comparitive analysis can be shown in between the three algorithms and also individually.