Abstract: The rapid growth of digital financial transactions has significantly increased the risk of fraudulent activities, making efficient fraud detection systems essential for modern financial institutions. Traditional Machine Learning models often struggle to accurately identify fraudulent transactions due to the class imbalance problem, where fraudulent instances are significantly fewer than legitimate ones. This paper proposes a cost-sensitive learning framework using Extreme Gradient Boosting (XGBoost) for financial fraud detection in imbalanced datasets. The proposed system incorporates data preprocessing, feature transformation, and cost-sensitive learning techniques to improve the detection of minority fraud cases. By assigning higher misclassification costs to fraudulent transactions, the model enhances its ability to detect fraud while maintaining overall performance. The system also provides real-time prediction through a user-friendly web interface, generating classification results along with a risk score for decision support. Experimental results demonstrate that the proposed approach achieves high accuracy, improved recall, and balanced performance, making it suitable for real-world financial fraud detection applications.

This work presented a financial fraud detection system based on a cost-sensitive XGBoost model to effectively address the class imbalance problem in transaction datasets. The proposed framework incorporates data preprocessing and feature transformation techniques to enhance data quality, followed by model training using cost-sensitive learning to improve the detection of minority fraudulent transactions. By assigning higher importance to fraud instances and optimizing hyperparameters, the model achieves improved recall and balanced classification performance compared to traditional approaches that do not consider data imbalance. Additionally, the use of an optimized decision threshold further enhances the model’s ability to detect fraud while maintaining a balance between false positives and false negatives. The system is implemented using a FastAPI-based backend, enabling real-time fraud prediction through a web-based interface. Comprehensive evaluation using metrics such as precision, recall, F1-score, and ROC-AUC demonstrates the effectiveness and reliability of the proposed approach. Overall, the framework provides a scalable and practical solution for detecting fraudulent activities in imbalanced financial datasets, supporting efficient decision-making in real-world applications. Future enhancements can include integration of advanced models, real-time data streaming, and adaptive learning techniques to further improve system performance and robustness.

[1]M. A. Alrasheedi, —Enhancing Fraud Detection in Credit Card Transactions: A Comparative Study of Machine Learning Models,‖ Computational Economics, pp. 1–20, 2025. [2]N. Damanik and C.-M. Liu, —Advanced Fraud Detection: Leveraging K-SMOTEENN and Stacking Ensemble,‖ IEEE Access, vol. 13, pp. 10358–10365, 2025. [3]Md. Alamin Talukder, Majdi Khalid, and Md. Ashraf Uddin, —An Integrated Multistage Ensemble Machine Learning Model for Fraudulent Transaction Detection,‖ Journal of Big Data, vol. 11, Article 168, pp. 1–25, 2024. [4]S. S. Suganya et al., —Ensemble Learning Approaches for Fraud Detection in Financial Transactions,‖ in Proceedings of the IEEE International Conference, pp. 1–6,2023. [5]E. Esenogho, I. D. Mienye, T. G. Swart, K. Aruleba, and G. Obaido, —A Neural Network Ensemble With Feature Engineering for Improved Credit Card Fraud Detection,‖ IEEE Access, vol. 10, pp. 16400–16407, 2022. [6]M. Alojail and S. Bhatia, —Novel Technique for Behavioral Analytics Using Ensemble Learning Algorithms in E-Commerce,‖ IEEE Access, vol. 8, pp. 150072–150080, 2020. [7]K. H. Ahmed, S. Axelsson, Y. Li, and A. M. Sagheer, —A credit card fraud detection approach based on ensemble Machine Learning classifier with hybrid data sampling,‖ Machine Learning with Applications, vol. 20, pp. 100675, 2025. [8]K. G. Dastidar, O. Caelen and M. Granitzer, —Machine Learning Methods for Credit Card Fraud Detection: A Survey,‖ IEEE Access, vol. 12, pp. 158939–158951, 2024. [9]D. Mienye, Y. Sun and Z. Wang, —Improved Credit Card Fraud Detection Based on Ensemble Learning,‖ IEEE Access, vol. 8, pp. 142048–142058, 2020. [10]R. Bin Sulaiman, V. Schetinin, and P. Sant, —Review of Machine Learning Approach on Credit Card Fraud Detection,‖Human-Centric Intelligent Systems, vol. 2, no. 1–2, pp. 55–68, 2022 [11]E. Ileberi, Y. Sun, and Z. Wang, —A Machine Learning Based Credit Card Fraud Detection using Genetic Algorithm,‖Journal of Big Data, vol. 9, Article 73, pp. 1–21, 2022. [12]R. Sailusha, V. Gnaneswar, R. Ramesh, and G. R. Rao, —Credit Card Fraud Detection Using Machine Learning,‖ Proceedings of the International Conference on Intelligent Computing and Control Systems (ICICCS), pp. 1264–1270, 2020 [13]D. Varmedja, M. Karanovic, S. Sladojevic, M. Arsenovic, and A. Anderla,—Credit Card Fraud Detection – Machine Learning Methods,‖Proceedings of the 18th International Symposium INFOTEH-JAHORINA, IEEE, pp. 1–6, 2019. [14]A. K. Singh, R. Kumar, and S. K. Singh, —Fraud Detection in Banking Transactions Using Machine Learning,‖Proceedings of the International Conference on Data Science and Engineering, Springer, pp. 210–220, 2022 [15]J. Jemai, A. Zarrad and A. Daud, —Identifying Fraudulent Credit Card Transactions Using Ensemble Learning,‖ IEEE Access, vol. 12, pp. 1–15, 2024. [16]R. Chhabra, S. Goswami and R. K. Ranjan, —A voting ensemble machine learning based credit card fraud detection using highly imbalanced data,‖ Multimedia Tools and Applications, vol. 83, no. 18, pp. 54729–54753, 2024. [17]X. Feng and S.-K. Kim, —Novel Machine Learning Based Credit Card Fraud Detection Systems,‖ Mathematics, vol. 12, no. 12, pp. 1–20, 2024.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close