Linking MOOC courseware to accommodate diverse learner backgrounds

Massive Open Online Courses (MOOCs) brings great opportunities to millions of learners. However, the size of the learner population and the heterogeneity of the learners' backgrounds make conventional one-size-fits-all pedagogy insufficient. For example, learners lacking in prior knowledge may struggle with different concepts. In this paper, we propose a framework educational content linking, to address the challenges. By linking and organizing scattered educational materials for a given MOOC into an easily accessible structure, this framework can provide guidance and recommendation of these contents, as well as improve navigation. Thus, learners can select appropriate supporting materials to suit their individualized needs and achieve self-exploring remediation. This paper describes an end-to-end case study, which found that learners, especially novices, can search learning materials faster without sacrificing accuracy, and can retain concepts more readily with our proposed approach. We have also obtained encouraging preliminary results that suggest that content linking can be achieved automatically using human language technology and stochastic modeling techniques.