A Collaborative Filtering Recommendation Based on User Profile Weight and Time Weight

Collaborative Filtering (CF) has proven to be the most widely used recommendation technology. However, the conventional CF ignores the impacts of user similarity caused by user profile, and it also can´t reflect the changes of user´s interests. To solve this problem, two weights are proposed: the user profile weight and the time weight. Also, the two weights are combined together and applied to a novel personalized recommendation system. The experimental results show that the improved method can obviously increase the recommendation precision.