Mining association rules in learning management systems

Learning management systems collect huge amounts of data that can later be analysed. The University of Rijeka uses MudRi e-learning system, which is based on the Moodle open source software. This paper focuses on the Programming course, for which data over several years are available. The data can be interpreted and valuable knowledge can be obtained and used for improving the quality of lectures, as well as making the lectures more suitable for students based on the actions and material deemed the most popular. Since the MudRi database contains many facts that might affect each other (e.g. homework might affect the final grade), association rule mining, which discovers regularities in data, is the most suitable data mining method. Apriori algorithm for the discovery of association rules is used for finding connections between various actions and final grades. Many interesting rules and information are discovered, which lead to conclusions on actions that seem to be in relation with the course success.