Designing Fault-Counter for Object-Oriented Software using Bagging Technique

Abstract

In Software Engineering, the faults present in software are the most critical issues, since they produce many incorrect and unreliable results. For developing reliable software, these faults must be resolved. In this project, a fault counter model is designed to predict the number of faulty modules present in a software project. There are four contributions in this work. The first contribution is to collect the dataset. The collected dataset contains numerous high-ranged and null values. In the second contribution, data pre-processing techniques like standard scaling and null-value removal are applied. The third contribution is to apply feature selection techniques to remove the least important features from the dataset. The fourth contribution is to predict the number of faults present in software projects using the Bagging Technique. The proposed model achieves a 0.55 R2_ Score.