LEARNING GROUP FORMATION FOR MASSIVE OPEN ONLINE COURSES (MOOCs)

Massive open online courses (MOOCs) describe platforms where users with completely different backgrounds subscribe to various courses on offer. MOOC forums and discussion boards offer learners a medium to communicate with each other and maximize their learning outcomes. However, oftentimes learners are hesitant to approach each other for different reasons (being shy, don’t know the right match, etc.). In collaborative learning contexts, the problem of automatic formation of effective groups becomes increasingly difficult due to very large base of users with different backgrounds. To address this concern, we propose an approach for group formation of users registered on MOOCs using a modified Particle Swarm Optimization (PSO) technique which automatically generates dynamic learning groups. The algorithm uses the profile attributes of users in terms of their age, gender, location, qualification, interests and grade as the grouping criteria. To form effective groups, we consider two important aspects: a) intra-group heterogeneity and b) inter-group homogeneity. While the former advocates the idea of diversity inside a particular group of users, the latter emphasizes that each group should be similar to one another. We test our algorithm on synthesized data sampled using the publicly available MITx-Harvardx dataset. Evaluation of the system is based on the fitness measures of groups generated using our algorithm which is compared against groups obtained using some of the standard clustering techniques like k-means. We see that our system performs better in terms of forming effective learning groups in the context of MOOCs.