Exploring Mathematical Problem-Solving Through Process Mining: Insights from Large-Scale Assessment Log Data

Abstract

In this study, we explore the application of process mining techniques on assessment log data to explore problem-solving strategies in Algebra. By analyzing sequences of student activities, we demonstrate the significant potential of process mining in identifying problem-solving strategies that lead to successful and unsuccessful outcomes. Our findings reveal that students who successfully solve the problem tend to follow one of three structured strategies, displaying a systematic process in filling the boxes of a Pascal’s triangle. Conversely, those who falter often start with a correct strategy but deviate by inserting incorrect values, especially in central boxes. Further analysis provides insights into the strategies and potential misconceptions among students from various disability groups. Notably, autistic students exhibit unique patterns, such as initiating the solution from the triangle’s right side, contrary to the common left-to-right strategy, and consistently applying this approach even when errors occur.