Citation :

Suman Basak, "Real-Time Fraud Detection Infrastructure: Building Sub 100ms Decision Engines with Streaming Analytics," International Journal of Computer Trends and Technology (IJCTT), vol. 74, no. 2, pp. 1-6, 2026. Crossref, https://doi.org/10.14445/22312803/IJCTT-V74I2P101

Abstract

Financial fraud detection must work extremely fast. Even a small delay of a few milliseconds can affect company revenue and frustrate users. This paper explains how we built a very fast fraud detection system that can handle millions of transactions every second and still make decisions in under 100 milliseconds. The system uses real-time data streaming instead of slow batch processing. It combines event streaming, real-time processing, fast in-memory data storage, and optimized machine-learning models to detect fraud instantly. Compared to traditional batch-based systems, this approach reduces processing time by 73% while still achieving very high accuracy (99.97%). We tested the system on real payment platforms such as digital wallets and peer-to-peer payment systems. According to the test result, the system can process about 2.4 billion transactions per month, with the majority of the decisions completed in under 80 milliseconds, even during peak load. Overall, this work provides both practical design guidance and performance insights for building fast, reliable, real-time fraud detection systems used in critical financial applications.

Keywords

Real-time fraud detection, Streaming analytics, Sub-millisecond latency, Apache Kafka, Apache Flink, Distributed systems, Machine learning inference, Payment systems.

References

[1] Nilson Report, Card Fraud Losses Reach $32.34 Billion, 2024. [Online]. Available: https://www.globenewswire.com/news-release/2022/12/22/2578877/0/en/Payment-Card-Fraud-Losses-Reach-32-34-Billion.html

[2] S. Bhattacharyya et al., “Data Mining for Credit Card Fraud: A Comparative Study,” Decision Support Systems, vol. 50, no. 3, pp. 602-613, 2011.
[CrossRef] [Google Scholar] [Publisher Link]

[3] Paris Carbone et al., “State Management in Apache Flink®: Consistent Stateful Distributed Stream Processing,” Proceedings of the VLDB Endowment, vol. 10, no. 12, pp. 1718-1729, 2017.
[CrossRef] [Google Scholar] [Publisher Link]

[4] Jay Kreps, Neha Narkhede, and Jun Rao, “Kafka: A Distributed Messaging System for Log Processing,” Proceedings of NetDB, pp. 1-7, 2011.
[Google Scholar] [Publisher Link]

[5] Jeffrey Dean, and Luiz André Barroso, “The Tail at Scale,” Communications of the ACM, vol. 56, no. 2, pp. 74-80, 2013.
[CrossRef] [Google Scholar] [Publisher Link]

[6] Daniel Crankshaw et al., “Clipper: A Low-Latency Online Prediction Serving System,” NSDI, pp. 613-627, 2017.
[Google Scholar] [Publisher Link]

[7] Tahereh Pourhabibi et al., “Fraud Detection: A Systematic Literature Review of Graph-Based Anomaly Detection,” Decision Support Systems, vol. 133, pp. 1-15, 2020.
[CrossRef] [Google Scholar] [Publisher Link]

[8] Johannes Jurgovsky et al., “Sequence Classification for Credit-Card Fraud Detection,” Expert Systems with Applications, vol. 100, pp. 234-245, 2018.
[CrossRef] [Google Scholar] [Publisher Link]

[9] Rafaël Van Belle, Bart Baesens, and Jochen De Weerdt, “CATCHM: A Novel Network-Based Credit Card Fraud Detection Method,” Decision Support Systems, vol. 164, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[10] Visa Advanced Authorisation: At work the world over, Visa Inc. [Online]. Available: https://corporate.visa.com/content/dam/VCOM/corporate/solutions/documents/visa-eu-advanced-authorization-case-study.pdf