MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621049628 A) filed by Dr. Pratibha Natthuji Atram; Dr. Deepali Ashish Deshmukh; Dr. Ranjana Kamaldas Jawanjal; Dr. Seema Madhukar Kale; Dr. Kalpana Laxmanrao Murade; Dr. Kavita Murlidhar Ingale; Dr. Parwati Keshavrao Shirke; Dr. Devidas Thosar; Dr. Pranjali Abhijit More; Ms. Kulkarni Amruta Ajay; Ms. Shweta Lilhare; and Ms. Swati Sucharita Barik on April 18, 2026, for Artificial Intelligence Based Digital Library Information Retrieval System With Semantic Search And Adaptive Knowledge Mapping Method.
Inventors include Dr. Pratibha Natthuji Atram; Dr. Deepali Ashish Deshmukh; Dr. Ranjana Kamaldas Jawanjal; Dr. Seema Madhukar Kale; Dr. Kalpana Laxmanrao Murade; Dr. Kavita Murlidhar Ingale; Dr. Parwati Keshavrao Shirke; Dr. Devidas Thosar; Dr. Pranjali Abhijit More; Ms. Kulkarni Amruta Ajay; Ms. Shweta Lilhare; and Ms. Swati Sucharita Barik.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The present invention discloses an artificial intelligence based digital library information retrieval system with semantic search and adaptive knowledge mapping, embodied as a machine-integrated device. The system comprises a processing unit operatively coupled with a memory unit, a storage unit containing digital content items, and an input interface configured to receive user queries in natural language. The processing unit is configured to transform the received queries into contextual vector representations using a trained artificial intelligence model and to compare the representations with encoded representations of stored digital content items maintained within a vector index. A similarity computation process is performed to generate relevance scores, followed by a ranking process to produce ordered retrieval results. The system further includes a knowledge mapping unit configured to construct and dynamically update a graph data structure representing relationships among content items, entities, and user interactions. The graph data structure is continuously refined based on user feedback to improve retrieval accuracy and contextual relevance.
Disclaimer: Curated by HT Syndication.
