Adaptive Socio-Recommender System for Open Corpus E-Learning

With the increase popularity of online education, the task of providing the right information to the right users has become a real challenge. Adaptive Hypermedia has been established as one solution to this challenge. However, many adaptation techniques are well suited for closed-corpus and are not compatible in open-corpus situation. Our solution for open-corpus adaptive navigation support is Social Adaptive Navigation Support (SANS). The most noticeable form of SANS could be offered through tracking visiting behavior of users, which we call traffic-based SANS. However, the fact that students visited a page does not mean that they found it useful. Therefore, we are also looking at annotation-based SANS in which students are encouraged to annotate the tutorial pages they visit. As part of my PhD research, I have developed a system that supports both traffic-based and annotation-based SANS. We have evaluated the system with two semesters of classroom studies. The result of the studies supports the idea of SANS and suggests more powerful navigation support for future work.