Time Series Analysis of VLE Activity Data

Virtual Learning Environments (VLE), such as Moodle, are purpose-built platforms in which teachers and students interact to exchange, review, and submit learning material and information. In this paper, we examine a complex VLE dataset from a large Irish university in an attempt to characterize student behavior with respect to deadlines and grades. We demonstrate that, by clustering activity profiles represented as time series using Dynamic Time Warping, we can uncover meaningful clusters of students exhibiting similar behaviors even in a sparsely-populated system. We use these clusters to identify distinct activity patterns among students, such as Procrastinators, Strugglers, and Experts. These patterns can provide us with an insight into the behavior of students, and ultimately help institutions to exploit deployed learning platforms so as to better structure their courses.