MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202631051574 A) filed by Mohammad Amir Khusru Akhtar; and Usha Martin University on April 22, 2026, for A Device-Implemented System And Method For Detecting Learning Difficulty And Adaptive Knowledge Modeling Using Interaction- Derived Behavioral Signals.
Inventors include Mohammad Amir Khusru Akhtar; Naghma Khatoon; Vinay Singh; Arvind Hans; and Bidisha Sarkhel.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: A device-implemented system and method for detecting learning difficulty and enabling adaptive knowledge modeling using interaction-derived behavioral signals are disclosed. The system captures user interaction data including keystroke dynamics, response latency, click behavior, dwell time, and revision patterns and transforms the data into numerical feature vectors. A cognitive state estimation module computes parameters including hesitation index, confusion score, and confidence decay derived from temporal interaction features. A dynamic graph representation models learners and concepts as nodes with relationships defined by interaction-derived metrics. A processing engine computes and dynamically updates relational interactions between nodes using a bounded and normalized computational model to identify weak knowledge regions. The system performs incremental graph updates and sparse pruning to achieve reduced computational complexity, improved convergence stability, and real-time processing with improved computational efficiency and system stability. An adaptive learning engine generates personalized learning pathways and predicts learning outcomes. The invention enables early detection of learning difficulty prior to explicit incorrect responses and supports scalable deployment across educational platforms.
Disclaimer: Curated by HT Syndication.
