Software Fault Detection using Multi-Distinguished-Features Sampling with Ensemble Random Forest Classifier

Abstract

Finding faults in software modules is an emerging issue in software reliability systems, and the assessment of the fault is performed by software fault prediction systems (SFPS). The identification process of fault-prone software modules is one of the prioritized aspects before initiating the testing process of the same modules. The SFPS helps improve software quality within the specified time and cost values. Early fault prediction in SFPS for the different software components showed significant results concerning the cost and time parameters. According to the state-of the art SFPS, ensemble-based classifiers were performed best and most cost-effective compared to other classifier methods. Recently, a random ensemble forest with adaptive synthetic sampling (E-RF-ADASYN) has been developed, is tested on a sample of PROMISE datasets, and shown the cost-effective classifier results. In the logistic regression to software quality models, and the other knowledge of account for prior probability and costs of misclassification. Probabilities and costs of misclassification in a logistic regression-based classification algorithm for software quality modeling. The decision tree algorithm is an ensemble learning approach for prediction. The algorithm works based on developing several decision trees and later decides the output class based on the most popular one. The proposed work focuses on developing an alternative sampling method called ensemble-random forest with multi-distinguishedfeatures sampling (E-RF-MDFS), for obtaining the best sample illustration for representing the entire dataset. Batinduced butterfly optimization (BBO) has been used for the feature extraction process. The experiments are conducted on 8 datasets of the PROMISE database. The proposed E-RF-MDFS has improved performance than E-RF-ADASYN in fault detection accuracy, real positive rate, and Pearson’s correlation coefficient. On comparing the performance of E-RF-based classifiers, the performance of the proposed MDFS is the best, with an FDA of 99.3 % (Xalan v2.6) than the ADASYN classifier