Improving the Scalability of MOOC Platforms with Automated, Dialogue-based Systems

摘要

In recent years e-learning environments got more and more popular, which resulted in a diverse, fast-growing community that among other anticipate individualisation and flexibility in means of customer support and recommendations. Scalability is required when the number of learners and the number of learning resources increase. This work targets dialogue-based systems used for Massive Open Online Course (MOOC) platforms to reduce human resources for answering technical support requests. The dialogue-based system must be available 24/7 and response immediately mainly to platform- specific frequently asked questions which were identified by an analysis of needs on existing support requests. Based on this comprehensive concepts, prototypes were implemented to prove the feasibility and explore the possibilities. The achievements of the deployment of a dialogue-based system for a MOOC platform were evaluated with well-defined metrics. These metrics showed, that even when deploying the most basic version of a dialogue-based system, the number of support requests created is reduced and thus the costly manual effort.