Collaborative filtering with automatic rating for recommendation
Mira Kwak, Dong-Sub Cho · 2002
In this paper, the authors describe a recommendation system designed to suggest new products to Web shopping mall customers. The recommender is meant to provide alternatives or new products that actively suit a user's tastes and meets his/her needs. Most previously proposed recommendation systems that use collaborative filtering can cause problems when there are insufficient user ratings. The authors address this problem by combining content-based filtering and collaborative filtering. Using an automatic rating method instead of a users' explicit rating, the inaccuracy of rating data is decreased.