What Should I Do Next? Adaptive Sequencing in the Context of Open Social Student Modeling

Abstract

One of the original goals of intelligent educational systems was to guide each student to the most appropriate educational content. In previous studies, we explored both knowledge-based and social guidance approaches and learned that each has a weak side. In the present work, we have explored the idea of combining social guidance with more traditional knowledge-based guidance systems in hopes of supporting more optimal content navigation. We propose a greedy sequencing approach aimed at maximizing each student’s level of knowledge and implemented it in the context of an open social student modeling interface. We performed a classroom study to examine the impact of this combined guidance approach. The results of our classroom study show that a greedy guidance approach positively affected students’ navigation, increased the speed of learning for strong students, and improved the overall performance of students, both within the system and through end-of-course assessments.