

Today, collaborative filtering techniques play a key role in many Web 2.0 applications. Currently, they are mainly used for business purposes such as product recommendation. Collaborative filtering also has potential for usage in “Social Semantic Web” e-learning applications in that the quality of a student provided solution can be heuristically determined by peers who review the solution, thus effectively disburdening the workload of teachers and tutors. This chapter presents a collaborative filtering algorithm which is specifically adapted for the requirements of e-learning applications. An empirical evaluation of the algorithm showed that the results of the collaborative filtering were more accurate than the self-assessment of the participants and that already four peer evaluations were generally enough to reach a satisfying accuracy. Based on these results, we developed a web based e-learning system (CITUC), which was successfully used in a university course in summer 2008. This chapter describes an evaluation of CITUC based on surveys, interviews and a detailed analysis of the system's usage by students. Our conclusion is that Social Semantic Web applications such as CITUC, which enable learners to review and comment on peer solutions, have high potential as a support for classic academic teaching in larger classes.