Detecting Knowledge Gaps from Conversational AI Interactions Using Curriculum Prerequisite Graphs
📰 ArXiv cs.AI
Detect knowledge gaps in students using conversational AI interactions and curriculum prerequisite graphs, improving personalized learning
Action Steps
- Build a prerequisite knowledge graph of course concepts using GPT-4
- Map student questions to curriculum topics using a few-shot text classifier
- Analyze interaction logs from conversational AI teaching assistants to identify knowledge gaps
- Configure a pipeline to integrate the text classifier and knowledge graph
- Test the pipeline using a dataset of question events, such as the 1,340 question events used in the evaluation
Who Needs to Know This
Educators and AI researchers can benefit from this approach to identify knowledge gaps and tailor their teaching methods, while developers can apply this to build more effective conversational AI systems
Key Insight
💡 Conversational AI interactions can be used as diagnostic signals to detect knowledge gaps in students
Share This
🤖 Identify knowledge gaps in students using conversational AI interactions & curriculum graphs! 📚
Key Takeaways
Detect knowledge gaps in students using conversational AI interactions and curriculum prerequisite graphs, improving personalized learning
Full Article
Title: Detecting Knowledge Gaps from Conversational AI Interactions Using Curriculum Prerequisite Graphs
Abstract:
arXiv:2606.10736v1 Announce Type: cross 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 f
Abstract:
arXiv:2606.10736v1 Announce Type: cross 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 f
DeepCamp AI