Recommending Personalized Review Materials Which Address A Student's Knowledge Gap

How can we Personalize Learning by leveraging recommendation systems? Personalized Learning (PL) is an approach to education that involves meeting each student at their current level of abilities and interests. PL is shown to significantly improve student achievement of learning outcomes. However, it is complex and expensive, and thus still inaccessible to many students. To improve access, I apply Recommendation Systems (RSs) (a class of applications that suggest items that might be of interest to a user) to personalize learning effectively and cost-efficiently. I do this by implementing a content-based filtering model that takes into consideration students’ mastery of concepts to recommend personalized review materials (RMs) that address the knowledge gap of a student undergoing a quantitative skills and reasoning (QR) program at Quest University Canada. A simulated dataset of five students and five RMs is used here for illustration, and the model is implemented as a fully functional web application accessible at http://personaltutor.ca. This model recommends RMs with the highest correspondence between the concepts the RM address and the concepts the learner needs to review. This result supports the thesis that universities can effectively and cost-efficiently personalize learning for students taking quantitative skills and reasoning programs by implementing a review materials recommendation system. Such systems can save students and teachers time and energy by helping them evaluate and find the most useful instructional materials. This and other benefits partially explain the growing interest in the use of RSs within the field of Technology Enhanced Learning (TEL)..