Software Defect Prediction Using Neural Network Based SMOTE

Rizal Broer Bahaweres, Fajar Agustian, Irman Hermadi, Arif Imam Suroso, Yandra Arkeman


Software defect prediction is a practical approach to improve the quality and efficiency of time and costs for software testing by focusing on defect modules. The defect prediction software dataset naturally has a class imbalance problem with very few defective modules compared to non-defective modules. Class imbalance can reduce performance from classification. In this study, we applied the Neural Networks Based Synthetic Minority Over-sampling Technique (SMOTE) to overcome class imbalances in the six NASA datasets. Neural Network based on SMOTE is a combination of Neural Network and SMOTE with each hyperparameters that are optimized using random search. The results use a nested 5-cross validation show increases Bal by 25.48% and Recall by 45.99% compared to the original Neural Network. We also compare the performance of Neural Network based SMOTE with SMOTE + Traditional Machine Learning Algorithm. The Neural Network based SMOTE takes first place in the average rank.


Software defect prediction; Class imbalance; Synthetic minority over-sampling technique; Neural network

Full Text: PDF


  • There are currently no refbacks.