MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202521112239 A) filed by Suja Sreejith Panickar Prerana Yadav; Vishal Nagare; Nia Uk; Rachit Udaseen; Rimjhim Sinha; Omkar Singh; and Vidhi Chawla, Pune, Maharashtra, on Nov. 16, 2025, for 'a system and method for automated fraud detection inupi transactions with real-time interpretability.'
Inventor(s) include Suja Sreejith Panickar; Prerana Yadav; Vishal Nagare; Nia Uk; Rachit Udaseen; Rimjhim Sinha; Omkar Singh; and Vidhi Chawla.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "Despite significant progress in digital transaction technologies, the challenge of real-time fraud detection in UPI systems remains largely unsolved due to the dynamic and high-velocity nature of financial data. Fraudulent transactions often mimic legitimate behaviour, occur in microseconds, and exploit small behavioural shifts that static systems fail to catch. To address these limitations, the invention provides an automated UPI fraud monitoring and interpretation system, which integrates multiple analytical modules into a unified, scalable framework. The system uses a hybrid detection approach combining XGBoost, LSTM, and Isolation Forest to detect a wide spectrum of fraud patterns, from structured anomalies to temporal shifts. It incorporates SMOTE and ADASYN for data balancing, PCA for dimensionality reduction, and SHAP for transparent explanation of model decisions."
Disclaimer: Curated by HT Syndication.
