This paper provides a service-oriented such solution which explore the ontology based modeling of users and documents in order to provide users with personalized recommendations of resources. Alongside with the adoption of semantic Web technologies for ontology based modeling, our approach aims a better pertinence for recommendations by adopting a hybrid recommendation technique, combining a collaborative filtering and a content-based approach. The collaborative filtering module adopts a Markov decision process in order to predict the next concept which will be focused by the user, tracking the user navigation through ontology instead through the structure of a particular site. The content based recommender module adopts a k nearest neighbors approach and exploits the similarities between user and document modeling.