MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122464 A) filed by Malla Reddy (MR) Deemed to be University; Malla Reddy University; Malla Reddy Engineering College For Women; Malla Reddy College Of Engineering And Technology; and Malla Reddy Vishwavidyapeeth, Malkajgiri, Telangana, on Dec. 5, 2025, for 'privacy-enhanced recommender system using federated learning.'
Inventor(s) include Dr. Kandru Arun Kumar; Dr. V. Swathi; Ms. Preethi Singireddy; Dr, Neeraja Vundyala; and Dr. Kanaka Durga Returi.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "This poses a privacy-preserving recommender system that makes use of federated learning for the purpose of training intelligent recommender systems without sending the user data to a data center. Each participating device or local node learns its own model parameters using its private data without having to share any other information with other devices and just by sending encrypted gradient updates to a central aggregator. This sort of mechanism where personal data cannot be accessed directly significantly mitigate the chances of the leakage of privacy and unauthorized profiling. The framework combines the differential privacy and secure aggregation methods to further protect the transmission of intermediate model updates. Independently interposed with these protections evolve the transmission of the decentralized optimization to make it possible to enhance model-aerobic collective improvement, fix sensitive details to the area that provide form. The architecture dynamically adapts to the variation of the distribution of the data or the user behavior or devices participation guarantees the consistency of the accuracy, even under heterogeneous conditions. A lightweight communication protocol is implemented for reduced synchronization delay and bandwidth consumption and therefore the model is suitable for large-scale deployment in mobile, enterprise, and IoT environments. The recommended system shows excellent quality of recommendations, faster convergence and privacy assurance as compared to conventional centralized recommender networks. It offers the first scalable basis for theoretically-privacy aware personalization for e-commerce, healthcare, and digital media services."
Disclaimer: Curated by HT Syndication.
