MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641050051 A) filed by Sr University, Warangal, Telangana, on April 20, for 'a hybrid human-ai system for real-time myth detection and depoliticization analysis in algorithmic content streams and synthetic environments.'
Inventor(s) include Mulupuri Pranathi; and Dr. Ritu 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 discloses a computer-implemented hybrid human-AI system for real-time detection and quantification of semantic abstraction patterns in multi-modal data streams. The system comprises an input acquisition module configured to receive textual, visual, and synthetic data, a preprocessing engine for generating structured feature representations, and a semantic decomposition engine for extracting multi-layer semantic features including literal, contextual, and abstracted representations. A trained classification module identifies semantic abstraction patterns, and a depoliticization scoring engine computes quantitative metrics including a context suppression index, naturalization score, and abstraction intensity level. A human-in-the-loop feedback module enables adaptive model refinement using reinforcement learning. An output interface generates real-time analytical outputs including classification labels, scores, and contextual explanations. The invention provides improved interpretability and computational detection of semantic transformation patterns in algorithmic systems, enabling scalable and efficient analysis across distributed computing environments."
Disclaimer: Curated by HT Syndication.
