An item-based weighted collaborative recommended algorithm based on screening users´ preferences

Facing the quality of the collaborative recommendation algorithm system, an item-based collaborative filtering recommendation algorithm based on screening users´ preferences was put forward. First, users have a unified interest measure for the project, and have a selection for the project based on the fuzzy membership of interests to build to project preference vector which reflects the user characteristics, using the cosine distance between vectors, the users were classified based on the weighted Item-based collaborative filtering. Experiments show that the algorithm is better than the traditional collaborative filtering algorithms in improving the recommendation dependability and accuracy.