A Methodology for Mining Data from Computer-Supported Learning Environments

Abstract

Computer-supported learning environments are usually adopted as platforms for distance-based education, but are also used as supporting tools for face-to-face educational settings. However, in such situations educators lose contact with their students and the way they access and use the content made available to them. This paper presents a methodology to process data collected from server logs and from the environments internal databases to provide feedback to authors and tutors about the content they offer. Two clustering algorithms, K-means and Self-Organizing Maps, were used to analyze the collected users’ interaction data and thus establish patterns of content access. An evaluation was performed with data collected from an actual environment used at a Brazilian university.