FR-CNN: SOFTWARE FAULT PREDICTION USING A NOVEL HYBRID HIERARCHICAL MODEL

(Received: 16-May-2026, Revised: 29-Jul.-2026 , Accepted: 19-Aug.-2026)

Authors Mehrasa M. Jouybari, Alireza Tajary, Mansoor Fateh, Esmaeel Tahanian,

Keywords #Software-fault prediction #Class imbalance #Deep learning #Ensemble learning #Machine learning #Hyper-parameters

Abstract Developing high-quality software products is directly related to the absence of unseen faults. The software fault prediction (SFP) methods have been considered to reduce the overall testing costs and discover probable faults. Various machine-learning and deep-learning algorithms have been used to predict faults. However, both deliver varying results regarding SFP. This research proposes a novel hybrid model (FR-CNN) based on a fine-tuned fully connected deep neural network, a random forest and a convolutional neural network. Harnessing the strengths of combining these techniques, the proposed model aims to enhance prediction performance while reducing overfitting, resulting in a robust framework. We conduct extensive empirical studies to illustrate the effectiveness of the proposed model in predicting faults on six projects of the BugHunter dataset. The empirical results demonstrate that FR-CNN achieves competitive fault-prediction performance while requiring lower computational complexity and training time than recent deep-learning approaches. In particular, it improves the best previously reported traditional machine-learning result on the BugHunter benchmark by up to 19.95% in F1- score.

References

[1] A. Arshad, S. Riaz, L. Jiao and A. Murthy, "Semi-supervised Deep Fuzzy C-mean Clustering for SoftwareFault Prediction," IEEE Access, vol. 6, pp. 25675–25685, 2018.

[2] F. Li, P. Yang, J. W. Keung, W. Hu, H. Luo and X. Yu, "Revisiting ‘Revisiting Supervised Methods forEffort‐aware Cross‐project Defect Prediction’," IET Software, vol. 17, no. 4, pp. 472–495, 2023.

[3] E. N. Akimova et al., "A Survey on Software Defect Prediction Using Deep Learning," Mathematics, vol.9, no. 11, p. 1180, 2021.

[4] S. K. Pandey et al., "Software Defect Prediction Using K‐PCA and Various Kernel‐based ExtremeLearning Machine: An Empirical Study," IET Software, vol. 14, no. 7, pp. 768–782, 2020.

[5] C. Jin, "Cross-project Software Defect Prediction Based on Domain Adaptation Learning andOptimization," Expert Systems with Applications, vol. 171, p. 114637, 2021.

[6] J. Pachouly, S. Ahirrao, K. Kotecha, G. Selvachandran and A. Abraham, "A Systematic Literature Reviewon Software Defect Prediction Using Artificial Intelligence: Datasets, Data Validation Methods, Approaches and Tools," Engineering Applications of Artificial Intelligence, vol. 111, p. 104773, 2022.

[7] Y. Hassouneh et al., "Boosted Whale Optimization Algorithm with Natural Selection Operators forSoftware Fault Prediction," IEEE Access, vol. 9, pp. 14239–14258, 2021.

[8] S. Kumar and S. S. Rathore, Software Fault Prediction: A Road Map, eBook ISBN978-981-10-8715-8,Springer, 2018.

[9] Z. Li, X.-Y. Jing and X. Zhu, "Progress on Approaches to Software Defect Prediction," Iet Software, vol.12, no. 3, pp. 161–175, 2018.

[10] X. Chen, Y. Zhao, Q. Wang and Z. Yuan, "MULTI: Multi-objective Effort-aware Just-in-time SoftwareDefect Prediction," Information and Software Technology, vol. 93, pp. 1–13, 2018.

[11] K. Wang, L. Liu, C. Yuan and Z. Wang, "Software Defect Prediction Model Based on LASSO–SVM,"Neural Computing and Applications, vol. 33, no. 14, pp. 8249–8259, 2021.

[12] S. K. Pandey, R. B. Mishra and A. K. Tripathi, "Machine Learning Based Methods for Software FaultPrediction: A Survey," Expert Systems with Applications, vol. 172, p. 114595, 2021.

[13] F. Khan et al., "Hyper-parameter Optimization of Classifiers, Using an Artificial Immune Network andIts Application to Software Bug Prediction," IEEE Access, vol. 8, pp. 20954–20964, 2020.

[14] C. Arun and C. Lakshmi, "Genetic Algorithm-based Oversampling Approach to Prune the ClassImbalance Issue in Software Defect Prediction," Soft Computing, vol. 26, no. 23, pp. 12915–12931, 2022.

[15] F. Matloob et al., "Software Defect Prediction Using Ensemble Learning: A Systematic LiteratureReview," IEEE Access, vol. 9, pp. 98754-98771, 2021.

[16] Ö. F. Arar and K. Ayan, "A Feature Dependent I Bayes Approach and Its Application to the SoftwareDefect Prediction Problem," Applied Soft Computing, vol. 59, pp. 197–209, 2017.

[17] M. Singh and D. S. Salaria, "Software Defect Prediction Tool Based on Neural Network," Int. J. ofComputer Applications, vol. 70, no. 22, 2013.

[18] Z. M. Zain et al., "Software Defect Prediction Harnessing on Multi 1-dimensional Convolutional NeuralNetwork Structure," Computers, Materials and Continua, vol. 71, no. 1, p. 1521, 2022.

[19] I. Goodfellow, Y. Bengio and A. Courville, Deep Learning, [Online], Available:http://www.deeplearningbook.org, MIT Press, 2016.

[20] M. M. Ahsan, T. E Alam, T. Trafalis and P. Huebner, "Deep MLP-CNN Model Using Mixed-data toDistinguish between COVID-19 and Non-COVID-19 Patients," Symmetry, vol. 12, no. 9, p. 1526, 2020.

[21] A. Abdu et al., "Semantic and Traditional Feature Fusion for Software Defect Prediction Using HybridDeep Learning Model," Scientific Reports, vol. 14, no. 1, p. 14771, 2024.

[22] S. Kaliraj, V. G. P. Sahasranth and V. Sivakumar, "A Holistic Approach to Software Fault Predictionwith Dynamic Classification," Automated Software Engineering, vol. 31, no. 2, p. 70, 2024.

[23] Q. Fei et al., "A Software Defect Prediction Method Using a Multivariate Heterogeneous Hybrid DeepLearning Algorithm," Computers, Materials, & Continua, vol. 82, no. 2, p. 3251, 2025.

[24] R. Malhotra and P. Singh, "DHG-BiGRU: Dual-attention based hierarchical gated BiGRU for softwaredefect prediction," Information and Software Technology, vol. 179, p. 107646, 2025.

[25] J. S. Shirabad and T. J. Menzies, "The PROMISE Repository of Software Engineering Databases," Schoolof Information Technology and Engineering, University of Ottawa, Canada, vol. 24, p. 3, 2005.

[26] M. Shepperd, Q. Song, Z. Sun and C. Mair, "Data Quality: Some Comments on the NASA SoftwareDefect Datasets," IEEE Transactions on Software Engineering, vol. 39, no. 9, pp. 1208–1215, 2013.

[27] E. N. Akimova et al., "PyTraceBugs: A Large Python Code Dataset for Supervised Machine Learning inSoftware Defect Prediction," Proc. of the 2021 28th Asia-Pacific Software Engineering Conf. (APSEC), pp. 141–151, Taipei, Taiwan, 2021.

[28] R. Ferenc, P. Gyimesi, G. Gyimesi, Z. Tóth and T. Gyimóthy, "An Automatically Created Novel BugDataset and Its Validation in Bug Prediction," J. of Systems and Software, vol. 169, p. 110691, 2020.

[29] S. T. Cynthia et al., "Feature Transformation for Improved Software Bug Detection Models," Proc. of the15th Innovations in Software Engineering Conf. (ISEC '22), Article no. 16, pp. 1–10, 2022.

[30] J. Jász, "The Effectiveness of Hidden Dependence Metrics in Bug Prediction," IEEE Access, vol. 12, pp.
77214–77225, 2024.

[31] S. S. Rathore and S. Kumar, "Linear and Non-linear Heterogeneous Ensemble Methods to Predict theNumber of Faults in Software Systems," Knowledge-based Systems, vol. 119, pp. 232–256, 2017.

[32] S. S. Rathore and S. Kumar, "Towards an Ensemble Based System for Predicting the Number of SoftwareFaults," Expert Systems with Applications, vol. 82, pp. 357–382, 2017.

[33] S. Wang and X. Yao, "Using Class Imbalance Learning for Software Defect Prediction," IEEETransactions on Reliability, vol. 62, no. 2, pp. 434–443, 2013.

[34] Z. Sun, Q. Song and X. Zhu, "Using Coding-based Ensemble Learning to Improve Software DefectPrediction," IEEE Trans. on Systems, Man and Cybernetics, Part C, vol. 42, no. 6, pp. 1806–1817, 2012.

[35] S. S. Rathore and S. Kumar, "An Empirical Study of Ensemble Techniques for Software FaultPrediction," Applied Intelligence, vol. 51, no. 6, pp. 3615–3644, 2021.

[36] S. Moustafa et al., "Software Bug Prediction Using Weighted Majority Voting Techniques," AlexandriaEngineering Journal, vol. 57, no. 4, pp. 2763–2774, 2018.

[37] I. H. Laradji, M. Alshayeb and L. Ghouti, "Software Defect Prediction Using Ensemble Learning onSelected Features," Information and Software Technology, vol. 58, pp. 388–402, 2015.

[38] T. Wang, Z. Zhang, X. Jing and L. Zhang, "Multiple Kernel Ensemble Learning for Software DefectPrediction," Automated Software Engineering, vol. 23, no. 4, pp. 569–590, 2016.

[39] S. K. Pandey et al., "BPDET: An Effective Software Bug Prediction Model Using Deep Representationand Ensemble Learning Techniques," Expert Systems with Applications, vol. 144, p. 113085, 2020.

[40] T.-Y. Yu, C.-Y. Huang and N. C. Fang, "Use of Deep Learning Model with Attention Mechanism forSoftware Fault Prediction," Proc. of the 2021 8th IEEE Int. Conf. on Dependable Systems and Their Applications (DSA), pp. 161–171, Yinchuan, China, 2021.

[41] H. Liang, Y. Yu, L. Jiang and Z. Xie, "Seml: A Semantic LSTM Model for Software Defect Prediction,"IEEE Access, vol. 7, pp. 83812–83824, 2019.

[42] M. Alenezi, M. Zagane and Y. Javed, "Efficient Deep Features Learning for Vulnerability DetectionUsing Character N-gram Embedding," Jordanian Journal of Computers and Information Technology (JJCIT), vol. 7, no. 01, pp. 25-38, 2021.

[43] U. Samal, "Hybrid Approach to Software Fault Prediction Using Particle Swarm Optimization and FuzzyTime Series," Quality and Reliability Engineering International, vol. 41, no. 5, pp. 2005–2018, 2025.

[44] H. Turabieh, M. Mafarja and X. Li, "Iterated Feature Selection Algorithms with Layered RecurrentNeural Network for Software Fault Prediction," Expert Systems with Appl., vol. 122, pp. 27–42, 2019.

[45] H. Ji et al., "A New Weighted I Bayes Method Based on Information Diffusion for Software DefectPrediction," Software Quality Journal, vol. 27, no. 3, pp. 923–968, 2019.

[46] K. Zhu et al., "Software Defect Prediction Based on Enhanced Metaheuristic Feature SelectionOptimization and a Hybrid Deep Neural Network," J. of Sys. and Software, vol. 180, p. 111026, 2021.

[47] B. S. Raghuwanshi and S. Shukla, "SMOTE Based Class-specific Extreme Learning Machine forImbalanced Learning," Knowledge-Based Systems, vol. 187, p. 104814, 2020.

[48] N. Nnamoko and I. Korkontzelos, "Efficient Treatment of Outliers and Class Imbalance for DiabetesPrediction," Artificial Intelligence in Medicine, vol. 104, p. 101815, 2020.

[49] N. Limsettho et al., "Cross Project Defect Prediction Using Class Distribution Estimation andOversampling," Information and Software Technology, vol. 100, pp. 87–102, 2018.

[50] O. Al Qasem, M. Akour and M. Alenezi, "The Influence of Deep Learning Algorithms Factors inSoftware Fault Prediction," IEEE Access, vol. 8, pp. 63945–63960, 2020.

[51] M. Adil, R. Ullah, S. Noor and N. Gohar, "Effect of Number of Neurons and Layers in an ArtificialNeural Network for Generalized Concrete Mix Design," Neural Computing and Appl., pp. 1–9, 2022.

[52] J. Zhang et al., "Random-forests-based Network Intrusion Detection Systems," IEEE Trans. on Systems,Man and Cybernetics, Part C (Applications and Reviews), vol. 38, no. 5, pp. 649–659, 2008.

[53] L. Qiao, X. Li, Q. Umer and P. Guo, "Deep Learning Based Software Defect Prediction,"Neurocomputing, vol. 385, pp. 100–110, 2020.

[54] A. B. Farid, E. M. Fathy, A. S. Eldin and L. A. Abd-Elmegid, "Software Defect Prediction Using HybridModel (CBIL) of Convolutional Neural Network (CNN) and Bidirectional Long Short-term Memory (Bi-LSTM)," PeerJ Computer Science, vol. 7, p. e739, 2021.