Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{191422,
        author = {Dipti Diliprao Mehare},
        title = {Credit Card Fraud Detection Using a Hybrid CNN–LSTM Deep Learning Architecture},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {8},
        pages = {6364-6368},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=191422},
        abstract = {The exponential growth of digital payment systems and e-commerce platforms has significantly increased the frequency and complexity of credit card fraud. Traditional machine learning approaches rely heavily on manual feature engineering and often fail to adapt to evolving fraud patterns and highly imbalanced transaction datasets. Deep learning models have emerged as powerful alternatives due to their capability to learn complex, nonlinear representations directly from data. However, individual deep learning models such as Convolutional Neural Networks (CNNs) or Long Short-Term Memory (LSTM) networks suffer from inherent limitations when applied independently. CNNs are effective in feature extraction but lack temporal awareness, while LSTMs model sequential behavior but may overlook feature-level interactions.
This paper proposes a hybrid CNN–LSTM deep learning architecture for credit card fraud detection that integrates spatial feature learning and temporal dependency modeling. The CNN component extracts discriminative feature representations from transaction attributes, while the LSTM component captures sequential transaction behavior over time. Extensive experiments conducted on benchmark credit card transaction datasets demonstrate that the proposed hybrid model outperforms traditional machine learning models and standalone deep learning architectures in terms of accuracy, recall, F1-score, and AUC. The results confirm the effectiveness of hybrid deep learning models for real-time, robust, and scalable fraud detection systems.},
        keywords = {Machine Learning, Deep Learning, CNN, LSTM, F1-score, AUC.},
        month = {January},
        }