Secure Flow: Detecting Fraudulent Payments In Upi Transactions Through Blockchain and Machine Learning Models
Main Article Content
Abstract
The proliferation of digital payment system has escalated vulnerabilities to sophisti-cated fraudulent activities. conventional detection methods often reliant on static rules exhibit limited efficiency against dynamic and novel thread patterns this research proposes a novel hybrid architecture term secure flow which Synergistically combines adaptive machine learning with blockchain technology to enhance the security transparency and adapt-ability of financial transaction monitoring The framework comprises of three core compo-nents continuously learning ML machine engine that evolves using real time data streams and immutable blockchain ledger for tramper proof audit Trail supplemented by smart contract driven alert protocols and efficient consensus algorithm designed specifically for throughput financial validations Empirical valuation conducted on Generated transaction data sets demonstrate significant precision recall and F1 Score matrix The Blockchain layer augment system resilience and auditability without imposing prohibitive computa-tional burdens consequently secure flow contributes to a more secure trustworthy digital payment environment by effectively mitigating fraud risk and fostering greater transac-tional confidence.