MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202611111362 A) filed by Dr Santosh Lal; M. Vijaya Bhaskara Reddy; Dr. Abhishek Dabre; Dr. S. V. Raut; Dr. Shailendra; Dr. K. Margandan; Revathi R; Kirubakaran D; Partha Ghosh; Bhawana Thakur Bargah; Mr Sushant Shivaji Kadam; and Mano Sethuramji. P on September 17, 2026, for World Model-Driven Intelligent Learning System For Real-Time Prediction And Autonomous Improvement Of Student Academic Performance.
Inventors include Dr Santosh Lal; M. Vijaya Bhaskara Reddy; Dr. Abhishek Dabre; Dr. S. V. Raut; Dr. Shailendra; Dr. K. Margandan; Revathi R; Kirubakaran D; Partha Ghosh; Bhawana Thakur Bargah; Mr Sushant Shivaji Kadam; and Mano Sethuramji. P.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention relates to a World Model-Driven Intelligent Learning System for Real-Time Prediction and Autonomous Improvement of Student Academic Performance. Existing educational platforms have increasingly adopted digital records, learning analytics, machine learning-based prediction, and student monitoring to understand academic performance. However, many such approaches mainly examine marks, attendance, activity levels, or individual learning events separately. They may provide a prediction of academic performance, but often do not maintain a continuously changing view of the student's learning condition or consider how a particular learning intervention may affect future performance. The present invention addresses this limitation by combining student data collection, learning-state assessment, a world model, real-time prediction, personalized intervention, and continuous feedback within a single system. The system collects current and historical learning information and generates a Learning Engagement Index (LEI), Academic Progress Score (APS), Knowledge Retention Index (KRI), Learning Consistency Factor (LCF), and Academic Risk Score (ARS). These measures are used to maintain an updated representation of the student's learning condition. A machine learning- based prediction module then estimates expected academic performance or possible academic risk. Based on the predicted condition, the system evaluates suitable learning actions and provides personalized recommendations. The results obtained after applying the recommendations are fed back into the system to update the learning state and improve later predictions and recommendations. The invention can support earlier identification of learning difficulties, more suitable academic interventions, timely alerts, improved student monitoring, and better academic decision-making, thereby enabling a learning system that can continuously adapt to individual student needs. FIG.1
Disclaimer: Curated by HT Syndication.
