Recommendation technology for help seeking in Massive Open Online Courses discussion forums implemented using a context-aware Matrix Factorization model to predict students' preferences for answering a given question. MOOC discussion forums need social recommendation which requires the match should be attractive in both directions , i.e., students looking for help could direct thier questions to capable peers and students should be able to find or given questions that match their interests. This recommendation framework allows for this two-way recommendation.
Yang, Diyi, et al. "Forum thread recommendation for massive open online courses." Educational Data Mining 2014. 2014.
Yang, Diyi, David Adamson, and Carolyn Penstein Rosé. "Question recommendation with constraints for massive open online courses." Proceedings of the 8th ACM Conference on Recommender systems. ACM, 2014.