Identifying Collaborative Problem-Solving Behaviors Using Sequential Pattern Mining

With the increasing adoption of collaborative learning approaches, instructors must understand students' problem-solving approaches during collaborative activities to better design their class. Among the multiple ways to reveal collaborative problem-solving processes, temporal submission patterns is one that is more scalable and generalizable in Computer Science education. In this paper, we provide a temporal analysis of a large dataset of students' submissions to collaborative learning assignments in an upper-level database course offered at a large public university. The log data was collected from an online assessment and learning system, containing the timestamps of each student's submissions to a problem on the collaborative assignment. Each submission was labeled as quick (Q), medium (M), or slow (S) based on its duration and whether it was shorter or longer than the 25th and 75th percentile. Sequential compacting and mining techniques were employed to identify pairs of transitions highly associated with one another. This preliminary research sheds light on the recurring submission patterns derived from the amount of time spent on each problem, warranting further examination on these patterns to unpack collaborative problem-solving behaviors. Our study demonstrates the potential of temporal analysis to identify meaningful problem-solving patterns based on log traces, which may help flag key moments and alert instructors to provide in-time scaffolding when students work on group assignments.