EUVIMEDEUVIMED entdecken
Crossref

A Machine Learning Based Approach for Predicting Credit Card Payment Fraud Transaction

Ajay M, Inbatamil R, Akash V, K. Gopal

International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)

DOI: 10.15662/ijeetr.2026.0802369 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von Ajay M und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Credit card fraud has become a critical concern in the digital payment ecosystem, with transaction volumes growing exponentially while fraudulent activities evolve in sophistication. Traditional rule-based and signature-based detection systems struggle to identify novel fraud patterns, particularly given the extreme class imbalance in transaction datasets (fraudulent transactions often represent less than 0.2% of all transactions). This paper presents a machine learning-based fraud detection framework that addresses these challenges through an integrated approach combining graph-based transaction modeling, contrastive learning for feature discrimination, and ensemble classification. The proposed system represents credit card transactions as graph networks, where card numbers and merchants form nodes, enabling effective capture of transaction relationships and temporal spending behavior. A Random Forest classifier serves as the primary detection engine, achieving 96.8% accuracy with a false positive rate of 2.1%. Contrastive learning enhances feature representations by maximizing separation between legitimate and fraudulent transaction patterns in the embedding space. The system is deployed as a web-based application using Flask framework with MySQL backend, providing real-time fraud prediction and location-based alert generation without requiring biometric authentication or additional hardware. Experimental evaluation on real-world credit card transaction datasets demonstrates that the proposed framework significantly outperforms traditional approaches, achieving high recall while maintaining precision. The system's scalable architecture and real-time alerting capability make it suitable for deployment in banking environments.

Zitieren

APA

M, A., R, I., V, A., & Gopal, K. (2026). A Machine Learning Based Approach for Predicting Credit Card Payment Fraud Transaction. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802369

Vancouver

M A, R I, V A, Gopal K. A Machine Learning Based Approach for Predicting Credit Card Payment Fraud Transaction. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802369

BibTeX
@article{M2026,
  author = {M, Ajay and R, Inbatamil and V, Akash and Gopal, K.},
  title = {A Machine Learning Based Approach for Predicting Credit Card Payment Fraud Transaction},
  journal = {International Journal of Engineering & Extended Technologies Research},
  year = {2026},
  volume = {08},
  number = {02},
  doi = {10.15662/ijeetr.2026.0802369}
}
Kostenlos starten →