MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202611077029 A) filed by Manipal University Jaipur on June 22, 2026, for A Cross-Domain Media Recommendation Method And System Using Engineered 'Mood' Features.
Inventors include Aditi Jha; Dr. Shivendra Dubey; Dr. Pranshu Pranjal; and Sakshi Dubey.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The present invention relates to a cross-domain media recommendation system for generating recommendations between different media domains including books and music. The system comprises a data acquisition module, a feature engineering module, a music processing module, a book processing module, a vectorization module, a similarity computation module, a recommendation generation module, and a user interface. The data acquisition module obtains music data and book data from respective sources. The feature engineering module generates a unified mood vocabulary to predefined mood categories. The music processing module derives mood features from audio attributes, while the book processing module derives mood features from descriptive tags associated with books. The vectorization module transforms the mood features into a numerical vector. The system trains vectorization models, and associated data structures for efficient recommendation generation and displays recommendations together with corresponding similarity scores through the user interface.
Disclaimer: Curated by HT Syndication.
