MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202621082028 A) filed by Symbiosis International Deemed University on July 02, 2026, for System And Method For Driver Behavior Rating Using Feature-Based Machine Learning And Llm- Driven Insights.
Inventors include Vaibhav Shirole; Dr. Aniket K. Shahade; and Priyanka V. Deshmukh.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: ABSTRACT SYSTEM AND METHOD FOR DRIVER BEHAVIOR RATING USING FEATURE-BASED MACHINE LEARNING AND LLM-DRIVEN INSIGHTS The present disclosure relates to a system (100) and method for rating driver behavior. The system (100) comprises an OBD-II data acquisition module (110), a pre-processing module (120), and a data annotation module (130) computing sub-scores including over-speed, rapid acceleration, cruising, harsh braking, consistency and sharp-turn. A composite score computation module (150) aggregates the sub-scores into a unified weighted composite driver score, which is predicted by a regression model bank (164) including a random forest regressor (166). An embedding generation module (180), a vector similarity index (182) using nearest- neighbor retrieval, a retrieval module (190) and a prompt construction module (200) cooperate with a large language model (210) to generate an interpretable driver behavior report through a report rendering module (220), enabling driver coaching, fleet management and usage-based insurance risk assessment. [
Disclaimer: Curated by HT Syndication.
