Detecting Knowledge Gaps from Conversational AI Interactions Using Curriculum Prerequisite Graphs
arXiv:2606.10736v1 Announce Type: new Abstract: Large online courses generate thousands of student questions directed at conversational AI teaching assistants, yet these interaction logs remain largely untapped as diagnostic signals. We present a pipeline that maps student questions from a conversational AI teaching assistant to curriculum topics using a few-shot text classifier, grounded in a GPT-4-extracted prerequisite knowledge graph of course concepts. Evaluated on 1,340 question events from 164 students in a graduate-level AI course, our classifier achieves 80.0% accuracy across 43 label
This research builds on the increasing prevalence of conversational AI in education and the growing need for effective assessment tools for large online learning environments.
This development offers a scalable method for diagnosing individual student knowledge gaps in large online courses, which can significantly improve educational outcomes and AI assistant efficacy.
The ability to accurately map student questions to curriculum topics using AI-generated prerequisite graphs changes how educators can leverage AI for personalized learning and intervention.
- · Online education platforms
- · Conversational AI developers
- · Students in large online courses
- · Educational technology sector
- · Traditional manual assessment methods
- · Inefficient AI teaching assistants
Improved educational outcomes and reduced instructor workload in large online learning environments.
Increased demand for AI-driven analytics and insights into student learning patterns and curriculum effectiveness.
The development of truly adaptive learning paths and personalized educational AI systems based on continuous, real-time knowledge gap identification.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL