MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202531129683 A) filed by Guru Nanak Institute Of Technology, Kolkata, West Bengal, on Dec. 20, 2025, for 'privacy-preserving credit card fraud detection utilizing federated model aggregation across disparate data distributions.'
Inventor(s) include Dr. Ananjan Maiti; Prof. Parameswar Das; Mr. Nirupam Saha; Dr. Sumit Das; and Mr. Ayan Dey.
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 disclosure relates to a privacy-preserving system and method for detecting credit card fraud using federated model aggregation across multiple financial institutions. Each institution independently trains a local machine learning model on its proprietary transaction data, while transmitting only model updates to a central aggregation server. The server consolidates updates into a global fraud detection model, which is iteratively redistributed to the participating institutions. The disclosure addresses class imbalance through statistical resampling and synthetic data techniques, and resolves heterogeneity across institutions through robust aggregation strategies. Secure communication and encryption protocols ensure compliance with data privacy regulations. The system achieves superior fraud detection accuracy without compromising sensitive data, offering a scalable and globally applicable framework for financial fraud prevention."
Disclaimer: Curated by HT Syndication.
