MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641085329 A) filed by P. L. Shailaja on July 12, 2026, for Machine Learning–based Adaptive Intelligent System For Real-Time Prediction, Optimization, Federated Learn-Based Privacy-Preserving.
Inventors include P. L. Shailaja, Assistant Professor, Department Of Cse Ai&ml, St. Peter'S Engineering College, Hyderabad; M. Supriya, Assistant Professor, Department Cse- Ai&ml, Geethanjali College Of Engineering And Technology, Cheeryala, Medchal, Telangana; Shaik Subhani, Assistant Professor, Department Of Cse; Dr. V. Sangeetha, Associate Professor, Department Of; Akula Nageswari, Assistant Professor, Department Of It; Neerukattu Siva Kumar, Assistant Professor, Department Of Cse Ai&ml, St. Peter'S Engineering College, Hyderabad, Telangana; and Dr. S. Rahamat Basha, Professor, Department Of Cse, St. Peter'S Engineering College, Hyderabad, Telangana.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: The present invention, Machine Learning–Based Adaptive Intelligent System for Real-Time Prediction, Optimization, and Federated Learning– Based Privacy Preservation, introduces an advanced distributed artificial intelligence framework designed to provide secure, adaptive, and efficient real-time prediction while preserving the privacy of sensitive information. The proposed system integrates adaptive machine learning algorithms, federated learning, intelligent optimization techniques, secure aggregation, differential privacy, and encrypted communication into a unified architecture that enables collaborative model training without transferring raw data. Each participating node independently preprocesses and trains local machine learning models using its own private dataset, after which only encrypted model parameters are securely transmitted to a centralized aggregation server. The server combines these updates to generate a robust global model that is redistributed for continuous learning and performance enhancement. The adaptive learning mechanism continuously updates model parameters based on newly available information, enabling accurate prediction and intelligent decision-making under dynamic operating conditions. The framework further incorporates communication- efficient optimization techniques, including parameter compression, adaptive synchronization, and selective model updates, to minimize network bandwidth requirements and computational overhead. Advanced privacy-preserving mechanisms protect sensitive information from unauthorized access while ensuring compliance with modern data protection regulations. The system supports multiple machine learning and deep learning architectures, including convolutional neural networks, recurrent neural networks, long short-term memory networks, transformer models, and ensemble learning techniques, making it suitable for diverse applications in healthcare, finance, agriculture, industrial automation, cybersecurity, transportation, smart cities, and Internet of Things (IoT) environments. The proposed invention provides a scalable, secure, communication-efficient, and intelligent solution for next-generation distributed artificial intelligence systems requiring real-time prediction, adaptive optimization, and privacy-preserving collaborative learning.
Disclaimer: Curated by HT Syndication.
