MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202521114764 A) filed by Symbiosis International, Pune, Maharashtra, on Nov. 20, 2025, for 'system and method for detecting fraudulent transactions in banking data using machine learning techniques.'
Inventor(s) include Krisha Shah; Pranjali Vishwakarma; Prasen Nikte; and Dr. Supriya Mahadevkar.
The application for the patent was published on Dec. 12, under issue no. 50/2025.
According to the abstract released by the Intellectual Property India: "A computer-implemented system for detecting fraudulent transactions in banking data using machine learning techniques is disclosed. The system comprises a processor coupled to a memory storing executable instructions for data preprocessing, feature engineering, and fraud detection operations. A data preprocessing module removes outliers and applies temporal feature extraction using cyclical transformations to decompose transaction timestamps into constituent components. A feature engineering module extracts geographical distances between cardholder and merchant locations using Haversine formula calculations and derives demographic features from user data. A machine learning module trains multiple supervised learning models and deep learning architectures including long short-term memory networks and autoencoders while addressing class imbalance through Synthetic Minority Over-sampling Technique. A fraud detection module analyses transaction patterns using trained models to generate fraud probability scores."
Disclaimer: Curated by HT Syndication.
