Software Fault Prediction using Inteligence Techniques

Abstract

Decision tree (DT) have been successfully applied for solving both classification and regression problems in many applications. This paper evaluates the capability of DT (Decision Tree) in predicting defect-prone software and compares its prediction performance against three intelligence technique in the context of PC1 dataset. we have used PC1 dataset (NASA dataset) which has sufficient parameters for analysis. As PC1 data is highly unbalanced data different balancing techniques have been applied. Ten-fold cross validation is performed throughout the study.