A Collaborative Filtering Recommendation Algorithm based on Domain Knowledge

Sparsity is one of the challenges in recommendation technologies. Traditional collaborative filtering usually evaluates user similarity based on intersection of users´ rating items, and it can not acquire accurate recommendation results when user rating data are extremely sparse. In order to eliminate the limitation above, a novel collaborative filtering algorithm based on domain ontology is presented: the method calculates similarity between items according to domain ontology, fills user rating matrix, and calculates users´ similarity with adjusted cosine measure. The experiment result shows that it can effectively improve recommendation quality even with extreme sparsity of user rating data.