Using feature selection and association rule mining to evaluate digital courseware

Abstract

Effective digital courseware should be easy to implement and integrate into instructional plans, saving teachers time and helping them support their students' learning needs. It should also not only enable students to achieve explicit learning objectives but also accelerate the pace at which they do so. This paper highlights the advantage of using Feature Selection techniques and Associative rule mining to get insightful knowledge from the log data from the Learning Management System (Moodle). The Machine Learning approach can be objectively deployed to obtain a predictive relationship and behavioral aspects that permits mapping the interaction behaviour of students with their course outcome. The knowledge discovered could immensely assist in evaluating and validating the various learning tools and activities within the course, thus, laying the groundwork for a more effective learning process. It is hoped that such knowledge would result in more effective courseware that provides for a rich, compelling, and interactive experience that will encourage repeated, prolonged, and self-motivated use.