Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM61849
Volume:
10 |
4 |
April 2026
Pages:
1--
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
independent researcher
PVG’s College of Science & Commerce
independent researcher
PVG’s College of Science & Commerce
independent researcher
PVG’s College of Science & Commerce
Abstract: The exponential growth of online financial transactions has significantly transformed the global digital economy. However, this rapid digitalization has simultaneously increased exposure to sophisticated fraudulent activities. Traditional rule-based fraud detection systems, which rely on predefined thresholds and static heuristics, struggle to adapt to evolving fraud strategies and large-scale transactional environments. As fraudsters continuously modify their techniques, adaptive and intelligent detection mechanisms become essential. Artificial Intelligence (AI), particularly Machine Learning (ML), offers data-driven approaches capable of identifying complex hidden patterns in transactional datasets. This shows a comprehensive study of AI-based fraud detection framework using models like supervised machine learning, specifically Logistic Regression and Random Forest. The Kaggle Credit Card Fraud Detection dataset, containing 284,807 transactions with severe class imbalance (only 0.172% fraud cases), is utilized for experimental evaluation. The proposed workflow consists of data preprocessing, imbalance handling, feature engineering, model training, and performance evaluation using Accuracy, Precision, Recall, F1-score, and ROC-AUC metrics. The findings indicate that while Logistic Regression provides interpreting and stable baseline performance, Random Forest significantly outperforms it in recall and ROC-AUC metrics, making it more efficient for fraud detection tasks where minimizing false negatives is critical. The study further identifies research gaps in explainable AI, concept drift adaptation, privacy-preserving learning, and real-time deployment scalability. The proposed framework contributes toward developing robust, scalable AI-driven fraud detection systems for online financial platforms.
KEYWORDS
In summary, Artificial Intelligence has great transformative potential in fraud detection for online transactions. Nevertheless, its application in practice demands a careful trade-off between performance, interpretability, scalability, and regulatory requirements. Future interdisciplinary research is imperative to develop next-generation fraud detection systems that can keep pace with the dynamic digital financial environment.
[1] A. Bhattacharyya, V. Jha, K. Tharakunnel, and J. C. Westland, “Data mining for credit card fraud: A comparative study,” Decision Support Systems, vol. 50, no. 3, pp. 602–613, 2011. [2] A. Dal Pozzolo, O. Caelen, R. A. Johnson, and G. Bontempi, “Calibrating probability with undersampling for unbalanced classification,” in IEEE Symposium Series on Computational Intelligence, 2015. [3] F. Carcillo, Y. Le Borgne, O. Caelen, and G. Bontempi, “Machine learning for financial fraud detection: A survey,” IEEE Access, vol. 9, pp. 1–20, 2021. [4] Y. Bengio et al., “Anomaly detection in financial transactions,” arXiv:1904.10604, 2019. [5] H. He and E. Garcia, “Learning from imbalanced data,” arXiv:1709.08920, 2017. [6] X. Zhang et al., “Graph-based fraud detection using relational learning,” arXiv:2409.13406, 2024. [7] Nature Scientific Reports, “Hybrid artificial intelligence for financial anomaly detection,” Sci. Rep., vol. 15, 2025, Art. no. s41598-025-00873-y. [8] U. Fiore et al., “Using SMOTE for credit card fraud detection,” Information Sciences, vol. 479, pp. 448–455, 2019. [9] C. Whitrow et al., “Transaction aggregation as a strategy for credit card fraud detection,” Data Mining and Knowledge Discovery, vol. 18, no. 1, pp. 30–55, 2009. [10] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. ACM SIGKDD, 2016. [11] S. Lundberg and S. Lee, “A unified approach to interpreting model predictions,” in Advances in Neural Information Processing Systems, 2017. [12] K. Bonawitz et al., “Federated learning for secure and privacy-preserving AI,” Proc. Machine Learning Systems, 2019.
Publication Date
4/29/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
4
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...