MUMBAI, India, March 13 -- Intellectual Property India has published a patent application (202621007167 A) filed by Priyanka V. Deshmukh; Kalyani P. Sable; Shrikant L. Satarkar; and Rupesh U. Pote, Amravati, Maharashtra, on Jan. 24, for 'a hybrid-adaptive artificial intelligence system for multilingual sentiment analysis of covid-19 social media data.'

Inventor(s) include Kalyani P. Sable; Shrikant L. Satarkar; Rupesh U. Pote; and Priyanka V. Deshmukh.

The application for the patent was published on March 13, under issue no. 11/2026.

According to the abstract released by the Intellectual Property India: "The present invention relates to a scalable, computer-implemented hybrid-adaptive system and method for multilingual sentiment analysis of real-time social media data, particularly COVID- 19 related content. The invention addresses challenges associated with linguistic diversity, high noise levels, contextual irrelevance, and rapidly evolving sentiment patterns in large-scale social media streams. The proposed system acquires multilingual social media data from online platforms and performs adaptive preprocessing comprising machine translation, semantic filtering, and noise reduction using transformer-based language models. Automated weak supervision is employed to generate sentiment labels through lexicon-based techniques, enabling large-scale training without manual annotation. Latent sentiment structures are identified using density-based clustering over contextual embeddings, while temporal sentiment trends are modeled to capture sentiment evolution over time. A hybrid deep learning architecture combining transformer-based classifiers with convolutional and recurrent neural networks performs sentiment classification with high accuracy. Further, a reinforcement learning module dynamically adapts model parameters in response to changes in data distribution, thereby improving robustness and temporal generalization. The invention provides improved accuracy, scalability, multilingual robustness, and adaptability over existing sentiment analysis systems, making it suitable for real-time public health surveillance, crisis management, and large-scale social media analytics during pandemics such as COVID-19."

Disclaimer: Curated by HT Syndication.