A Benchmarking Study of K-Means and SOM Approaches Applied to a Set of Features of MOOC Participants

Abstract

MOOC format is characterized by the great diversity of enrolled people. This heterogeneity of participants represents a challenging opportunity in order to identify underlying relationships in the internal structure of features that make up participants’ profiles. This paper has the aim of identifying and analyzing a feasible set of MOOC participants’ profiles with the use of two unsupervised clustering techniques, K-Means as a partitional clustering algorithm and Kohonen’s Self-Organizing Maps (SOMs), hereinafter SOM, as a representative technique of Artificial Neural Networks (ANNs).