A content-based collaborative recommender system with detailed use of evaluations

We present a hybrid recommender model that combines the benefits of both content-based filtering and collaborative filtering. In this model, each document profile is represented as a pair of a keyword vector and an evaluation vector. Each user profile, on the other hand, is represented as a matrix of dependency values in relation to other users according to each keyword. This type of recommender system can provide more appropriate documents to suit a user´s personal information need. The simulation results showed that our model can provide appropriate documents to users with higher precision than other non-hybrid information filtering models