A contextualized, differential sequence mining method to derive students' learning behavior patterns

Abstract

Computer-based learning environments can produce a wealth of data on student learning interactions. This paper presents an exploratory data mining methodology for assessing and comparing students' learning behaviors from these interaction traces. The core algorithm employs a novel combination of sequence mining techniques to identify dierentially frequent patterns between groups of students (e.g., experimental versus control conditions or high versus low performers). We extend this technique by contextualizing the sequence mining with information about the student's performance over the course of the learning interactions. Specically, we employ a piecewise linear segmentation algorithm in concert with the dierential sequence mining technique to identify and compare segments of students' productive and unproductive learning behaviors. We present the results from the application of this exploratory data mining methodology to learning interaction trace data gathered during a recent middle school class study with the Betty's Brain learning environment. These results illustrate the potential of this methodology in identifying learning behavior patterns relevant to the investigation of metacognition and strategy use.