MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641064004 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on May 21, for 'a fraud detection system for financial transaction networks.'
Inventor(s) include Charanya; and Manisha RB.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to a system and method for detecting fraud in financial transaction networks is disclosed, wherein transaction data including amount, timestamp, device, location, merchant, and account history is ingested from multiple financial platforms and used to construct a continuously updated dynamic causal transaction graph representing entities as nodes and transactions as edges. Suspicious transactions are identified as fraud seed events using anomaly detection techniques based on abnormal patterns such as unusual amounts, geographic changes, device usage, and rapid transaction sequences. A fraud influence gradient engine evaluates the direction and intensity of fraud propagation across interconnected nodes using connection strength, risk variation, and temporal decay, while a node resistance module computes resistance scores based on behavioral consistency, interaction stability, and temporal reliability. A temporal fraud propagation resistance scoring engine combines propagated fraud influence with node resistance to generate a fraud risk score, enabling detection of coordinated fraud activities. A decision module classifies transactions or nodes as legitimate, suspicious, or fraudulent and outputs corresponding approval, verification, or blocking actions, with the system operating in near real time through continuous graph updates and recomputation of risk metrics."
Disclaimer: Curated by HT Syndication.
