A Clustering Methodology of Web Log Data for Learning Management Systems.
Year: 2012
Type: article
Abstract: Learning Management Systems (LMS) collect large amounts of data. Data mining techniques can be applied to analyse their web data log files. The instructors may use this data for assessing and measuring their courses. In this respect, we have proposed a methodology for analysing LMS courses and students ’ activity. This methodology uses a Markov CLustering (MCL) algorithm for clustering the students ’ activity and a SimpleKMeans algorithm for clustering the courses. Additionally we provide a visualisation of the results using scatter plots and 3D graphs. We propose specific metrics for the assessment of the courses based on the course usage. These metrics applied to data originated from the LMS log files of the Information Management Department of the TEI of Kavala. The results show that these metrics, if combined properly, can quantify quality characteristics of the courses. Furthermore, the application of the MCL algorithm to students ’ activities provides useful insights to their usage of the LMS platform.
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W2100780494