Evaluation of e-learning contents based on semantic social network and measurement of their quality values

The rapidly development of Internet technologies cause to increase the size of the data on the Internet and to become difficult access to quality content. Within this data complexity, the new approaches are needed to bring the desired information. The combination of recommender systems and semantic web technologies has made significant improvements for information filtering, especially in social networks. In this study, the hybrid recommendation methods have been developed by using semantic web technologies to solve the chronic problems of recommender systems such as cold start, data sparsity, scalability. Using these methods in the domain of e-learning, the quality values of the contents have been estimated for the learner. In this thesis, three ontologies have been designed for the learner, the learning object and the learning domain. The hybrid methods have been developed using content-based and collaborative filtering. Then, the hybrid methods have been combined and the output set has been improved according to the learning path. The quality values of the learning objects have been obtained using the output set. The hybrid recommendation method developed in this study, has been used to measure the quality of e-learning contents for learners.