MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202611027422 A) filed by Manipal University, Jaipur, Rajasthan, on March 9, for 'mood-based cross-domain recommender system: emotion-aware uplifting recommendations across songs, movies, and books.'
Inventor(s) include Dr. Shivendra Dubey; Bhanuni Sunkara; Ponnada Divesh; Sakshi Dubey; and Dr. Rahul Sharma.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to a mood-based cross-domain recommender system that is qualified to promote mood-uplifting recommendations covering songs, books, and movies. The system comprises a data acquisition module to collect emotional labels from users for music, movies, and books; a data preprocessing module to produce a uniform set of mood labels for all domains, including happy, sad, calm, etc; an emotional detection model is used to predict user emotion using input that is given via text; a feature extraction module extract textual, audio and video features; a cross domain alignment module to construct an affective knowledge graph helpful in linking similar emotions across different domains; training module to train the model; and an output module to predict the top N recommendations as per the domain, resulting in uplifting the mood. This system explores real-time emotions of users using Natural Language Processing (NLP) to propose content that helps in elevating users' emotional status. The Integration of hybrid recommender systems/techniques blends content-based filtering, collaborative filtering, and deep learning (similarity models), making this idea as unique as possible. The present system offers users contextually relevant, mood-uplifting recommendations, contributing to a more empathetic and intuitive digital experience throughout the cross-domain entertainment."
Disclaimer: Curated by HT Syndication.
