Interaction Networks: Generating High Level Hints Based on Network Community Clusterings

We introduce a novel data structure, the Interaction Network, for representing interaction-data from open problem solving environment tutors. We show how using network community detecting techniques are used to identify subgoals in problems in a logic tutor. We then use those community structures to generate high level hints between subgoals. The preliminary results show that using network analysis techniques are promising for exploring and understanding user data from open problem solving environments.