Towards a Better Graphlet-based Mind Map Metric for Automating Student Feedback

In this work, we design a new metric that helps analyze the difference between a student’s mind map and an expert’s mind map. The goal is to determine if student’s are learning and to provide them with automatic feedback on what they are not understanding. In previous work, we used various graph analysis metrics applied to the mind map to see if there is evidence of learning, and in this work, we combine ideas from two of the most successful of these metrics by creating a new tool that checks if small sub-graphs exist in both a student and the criterion map (an experts mind map). By analyzing the results of these matches, we create a global metric that we then compare to our previous metrics and find that this new metric has similar behavior. This is positive since this metric provides a means for more interesting feedback to students.