Abstract

Collaboration Filter is one of well-known effective methods for recommendation. The suggestions are based on the mass of user ratings for various items, which are used as explicit feedback. However, implicit feedback such as the time spent on website, how soon the user skipped the song, and the sequence of selected items have also proved to be useful in recommender systems. In this paper, we proposed a hybrid recommendation algorithm considering both explicit and implicit feedback to produce better recommended list. An iteration process is proposed for using implicit feedback in order to find effective feedback for recommendation. Time window is also used to limit the impact range of implicit feedback. Experimental results on MovieLens datasets show that the implicit feedback can effectively compensate for the shortcomings of explicit feedback and the proposed algorithm is more accurate than the traditional one.

Paper Details