Context-Aware Fraud Detection in Bank-to-Crypto On-Ramp Transactions Using Hybrid Machine Learning

DOI

10.2991/978-94-6239-616-6_47How to use a DOI?

Keywords

Cryptocurrency laundering; Off-ramp transaction monitoring; Hybrid machine learning; Anomaly detection; AntiMoney Laundering (AML); Financial fraud detection; Identity forgery detection

Abstract

The rapid expansion of digital payments and cryptocurrencies has created both opportunities and vulnerabilities in the modern financial landscape. A critical concern emerges when funds are swiftly transferred from traditional banking systems to cryptocurrency wallets, often using fraudulent or compromised accounts. These off-ramp transactions are challenging to identify through conventional mechanisms, exposing major gaps in existing fraud prevention frameworks. Current detection strategies, predominantly rule-based or dependent on isolated machine learning models, can capture simple irregularities but fail to adapt to complex and evolving fraud patterns. Consequently, they suffer from low detection accuracy, delayed identification, and susceptibility to large-scale scams. To mitigate these issues, this paper introduces an integrated fraud detection framework that fuses traditional banking transaction data with cryptocurrency wallet activity using a hybrid machine learning approach. The proposed system leverages Bidirectional Long Short Term Memory (BiLSTM) networks for sequential transaction analysis, Random Forest (RF) for feature-based classification, and MultiLayer Perceptron (MLP) for modeling non-linear dependencies. Additionally, the YOLO algorithm is integrated to identify forged or duplicate identities in real time. The inclusion of anomaly detection and continuous monitoring enhances adaptability and responsiveness. This combined approach significantly improves detection precision, minimizes false positives, and offers financial institutions a stronger defense mechanism against crypto-related fraud.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - P. Karthikeyan
AU  - S. Geetha
AU  - K. Thulasie
AU  - Niha Nafiza Shafi
AU  - S. Jeyvanti
PY  - 2026
DA  - 2026/03/31
TI  - Context-Aware Fraud Detection in Bank-to-Crypto On-Ramp Transactions Using Hybrid Machine Learning
BT  - Proceedings of the International Conference on Artificial Intelligence and Secure Data Analytics (ICAISDA 2025)
PB  - Atlantis Press
SP  - 628
EP  - 640
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-616-6_47
DO  - 10.2991/978-94-6239-616-6_47
ID  - Karthikeyan2026
ER  -