Presentation Information
[5O3-IS-5b-05]Supporting Goal Maintenance in Digital LearningGoal Maintenance by Integrating Semantic Analysis and Cognitive Modeling
〇Nilupul Heshan Randika Kodikara1, Umito Sorita1, Nethmee Rasanga gunasekara1, Joy Karmoker1, Junya Morita1 (1. Shizuoka University)
regular
Keywords:
Intelligent tutoring,cognitive modeling,Motivation Support,online learning
As digital learning environments grow more complex, learners face information overload that disrupts autonomy, which is essential for tracking learning goals that need to be achieved. To address these challenges, we propose a system that identifies such disruptions and supports goal maintenance by monitoring learner behavior, estimating interests, and assessing the relevance of observed content to the original goals. System has three modules: Semantic analysis module, which leverages language models to evaluate semantic relevance between learners' original goals and browsing behavior, enabling the detection of goal drift. Cognitive reading module, tracking real-time information-seeking behaviors based on a personalized cognitive model that incorporates users' past experiences and psychological parameters estimated by pre-executed test batteries. Model interprets behaviors such as scrolling and navigation in real time to represent the internal processes of the learner. Predictive machine learning module, integrating semantic and behavioral features to forecast outcomes such as distraction, task completion time, interest fluctuations. We have developed these modules independently, and future work will focus on integration. In this presentation, we report preliminary experiments on the semantic analysis, demonstrating the feasibility of tracking goal maintenance.