Mining Frequent Learning Pathways from a Large Educational Dataset

TL;DR

This paper introduces a novel approach to mining frequent learning pathways from large educational datasets by combining sequence clustering and graph-based process discovery, revealing common student learning routes.

Contribution

The paper presents a new sequence clustering algorithm and a graph-based process discovery method tailored for large-scale educational data analysis.

Findings

Identified common learning pathways among students

Improved process mining results through clustering

Revealed high-frequency sequences of learning activities

Abstract

In this paper, we describe data mining techniques used to extract frequent learning pathways from a large educational dataset. These pathways were extracted as a directed graph that encoded student learning processes. Our dataset contains more than 800 million interactions of over 3 million anonymized students in an online learning platform. Performing process mining on large and complex datasets regularly yields incomprehensible process models. Although, if we cluster data and obtain groups following similar processes, we can greatly improve process mining results. To this end, we developed a sequence clustering algorithm that let us group students who followed similar learning pathways. To extract frequent learning pathways from these clusters of data, we developed a graph-based process discovery algorithm that revealed to us the sequences of learning activities that many students followed. These sequences represented highways of student learning.