A UNIFY METHOD BETWEEN COLLABORATIVE FILTERING AND CONTENT-BASED FILTERING BASED ON GRAPH MODEL

Abstract

Recommender systems are the capable systems of providing appropriate information and removing unappropriate information for Internet users. The recommender systems are built based on two main information filtering techniques: Collaborative filtering and content-based filtering. Each method exploits particular aspects related to content features or product usage habit of users in the past to predict a brief list of the most suitable products with each user. Content-based filtering perform effectively on documents representing as text but have problems selecting information features on multimedia data. Collaborative filtering perform well on all  information formatsbut have problems with sparse data and new users. In this paper, we propose a new unify method between collaborative filtering and content-based filtering based on graph model. The model allows us to shift general hybrid filtering recommender problem to collaborative filtering recommender problem, then build  new similar measures based on graph to determine similarities between two users or two items, these similar measures are used to predict suitable products for users in the system. The experimental results on real data sets about films show that the proposed methods utilize advantages effectively and are disadvantages significant limitations of basedline methods.