Detecting Credit Card Fraud Using a Hybrid CNN-RNN Model
DOI:
https://doi.org/10.21070/jicte.v9i2.1680Keywords:
Deep learning, Fraud detection, Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), LSTMAbstract
Credit card fraud detection represents a pressing challenge due to the rarity and evolving nature of fraudulent transactions. This study suggests a hybrid deep learning framework combining Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), specifically Long Short-Term Memory (LSTM) units, applied to the 2013 European cardholder transactions dataset released on Kaggle. A publicly accessible dataset of actual credit card transactions is used to train and evaluate the model, and class imbalance is addressed by using the Synthetic Minority Over-sampling Technique (SMOTE). Experimental results demonstrate the Outstanding performance of the proposed Model that a CNN front-end is effective in extracting local transaction patterns, while RNN layers model sequential dependencies within transaction sequences.Outperform traditional machine learning baselines including Logistic Regression, Random Forest, and XGBoost as well as single deep learning models. Reported performance metrics for this hybrid model include precision (up to 99.4%), recall (up to 99.9%), F1-score (up to 99.7%), accuracy (up to 99.7%), and ROC-AUC (up to 0.999) on the Kaggle dataset .However, most studies rely on random or unspecified data splits and emphasize ROC-AUC or accuracy metrics rather than class imbalance aware measures such as Precision-Recall AUC; time-aware evaluation procedures are rarely detailed. These findings suggest that hybrid CNN-RNN models hold significant promise for credit card fraud detection, while underscoring the need for more rigorous evaluation methodologies and transparent reporting of model architecture and metrics.
Highlights:
-
Hybrid Model: CNN extracts local transaction patterns, RNN (LSTM) captures sequential dependencies.
-
High Performance: Achieves precision 99.4%, recall 99.9%, F1-score 99.7%, accuracy 99.7%, ROC-AUC 0.999.
-
Research Gap: Most studies ignore time-aware evaluation and Precision-Recall AUC despite class imbalance.
Keywords: Deep learning, Fraud detection, Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), LSTM.
References
[1] G. Yoganandham and G. Elanchezhian, “Analyzing the Economic Impact of Credit Card Fraud: Activation, Limit Upgrades, Cashback Scams, Discount Fraud, and Overdraft Risks,” Degrés, vol. 9, no. 11, 2024
[2] V. Ramaiya, P. Goyal, and N. K. Dubey, “Cybersecurity Threats and Trust Dynamics in Digital Payment Systems: An Analysis of Domestic Fraud and User Perspectives,” in Proc. Int. Conf. on Advancements in Smart Computing and Information Security, Cham, Switzerland: Springer Nature Switzerland, Oct. 2024, pp. 116–138.
[3] R. Udayakumar, A. Joshi, S. S. Boomiga, and R. Sugumar, “Deep fraud Net: A deep learning approach for cyber security and financial fraud detection and classification,” Journal of Internet Services and Information Security, vol. 13, no. 3, pp. 138–157, 2023.
[4] N. Ibrahim, A. Abbas, and F. Khorsheed, “A systematic review for misuses attack detection based on data mining in NFV,” Sakarya University Journal of Computer and Information Sciences, vol. 6, no. 3, pp. 239–252, 2023, doi: 10.35377/saucis.1379047.
[5] R. K. Kennedy, F. Villanustre, T. M. Khoshgoftaar, and Z. Salekshahrezaee, “Synthesizing class labels for highly imbalanced credit card fraud detection data,” Journal of Big Data, vol. 11, no. 1, p. 38, 2024 .
[6] D. O. Njoku, V. C. Iwuchukwu, J. E. Jibiri, C. T. Ikwuazom, C. I. Ofoegbu, and F. O. Nwokoma, “Machine learning approach for fraud detection system in financial institution: A web base application,” Machine Learning, vol. 20, no. 4, pp. 01–12, 2024.
[7] L. C. de Jesus Jr, F. Fernández-Navarro, and M. Carbonero-Ruz, “Enhancing financial time series forecasting through topological data analysis,” Neural Computing and Applications, vol. 37, no. 9, pp. 6527–6545, 2025.
[8] S. F. Ahmed, M. S. B. Alam, M. Hassan, M. R. Rozbu, T. Ishtiak, N. Rafa, et al., “Deep learning modelling techniques: current progress, applications, advantages, and challenges,” Artificial Intelligence Review, vol. 56, no. 11, pp. 13521–13617, 2023.
[9] I. D. Mienye, T. G. Swart, and G. Obaido, "Recurrent neural networks: A comprehensive review of architectures, variants, and applications," Information, vol. 15, no. 9, p. 517, 2024.
[10] F. M. Salem, "Recurrent neural networks (RNN)," in Recurrent Neural Networks: From Simple to Gated Architectures, Cham, Switzerland: Springer International Publishing, 2021, pp. 43–67.
[11] L. Alzubaidi, J. Zhang, A. J. Humaidi, A. Al-Dujaili, Y. Duan, O. Al-Shamma, et al., "Review of deep learning: concepts, CNN architectures, challenges, applications, future directions," Journal of Big Data, vol. 8, no. 1, p. 53, 2021.
[12] B. Wiese and C. Omlin, "Credit card transactions, fraud detection, and machine learning: Modelling time with LSTM recurrent neural networks," in Innovations in Neural Information Paradigms and Applications, Berlin, Heidelberg: Springer Berlin Heidelberg, 2009, pp. 231–268.
[13] https://www.kaggle.com/datasets/mlg-ulb/creditcardfraud
[14] S. Janbhasha, C. S. Kumar, V. Sitharamulu, B. M. Babu, H. R. Battu, and K. Venkataramana, "A hybrid approach for fraud detection in digital wallet transactions using adversarial autoencoders and gated recurrent units," Engineering, Technology & Applied Science Research, vol. 15, no. 4, pp. 25532–25537, 2025.
[15] I. Benchaji, S. Douzi, B. El Ouahidi, and J. Jaafari, "Enhanced credit card fraud detection based on attention mechanism and LSTM deep model," Journal of Big Data, vol. 8, no. 1, p. 151, 2021.
[16] J. Karthika and A. Senthilselvi, "Detection of credit card fraud detection using HPO with inception based deep learning model," in Proc. 2023 5th Int. Conf. Inventive Res. Comput. Appl. (ICIRCA), Aug. 2023, pp. 70–77.
[17] E. Ileberi and Y. Sun, "A Hybrid Deep Learning Ensemble Model for Credit Card Fraud Detection," IEEE Access, 2024.
[18] S. Sharma, A. Kataria, J. K. Sandhu, and K. R. Ramkumar, "Credit card fraud detection using machine and deep learning techniques," in Proc. 2022 3rd Int. Conf. for Emerging Technology (INCET), May 2022, pp. 1–7.
[19] S. El Kafhali, M. Tayebi, and H. Sulimani, "An optimized deep learning approach for detecting fraudulent transactions," Information, vol. 15, no. 4, p. 227, 2024.
[20] D. Elreedy, A. F. Atiya, and F. Kamalov, "A theoretical distribution analysis of synthetic minority oversampling technique (SMOTE) for imbalanced learning," Machine Learning, vol. 113, no. 7, pp. 4903–4923, 2024.
Downloads
Published
License
Copyright (c) 2025 Zainab Hassan Mohammed, Nebras Jalel Ibrahim, Ahmed K. Abbas
This work is licensed under a Creative Commons Attribution 4.0 International License.
