A Complete Chatbot based Architecture for answering user's Course-related queries in MOOC platforms

Abstract

MOOCs, after first being introduced in 2008, have since drawn attention around the world for their advantages and also criticism of their drawbacks. Interactivity with the instructor and personalized experience while learning are some of the main aspects in which the current MOOCs show scope of improvement. In this paper, we are proposing a novel chatbot architecture that can act as teaching assistance to answer queries faced by learners in MOOCs. The chatbot will be trained on the course material using the popular ALBERT model to develop its knowledge base. The chatbot will answer a large number of student's queries, thus reducing the workload of the instructor(s) to answer all the queries. This will open a new avenue for instructor-chatbot-learner interaction, where all of these three will compound each other's value.