Abstract

Personalized recommendation technology provides users with more rapid and effective information acquisition channels. The existing recommendation algorithms that focus on recommendation accuracy will misguide users to a few hot commodities, thus creating many long-tail commodities. As a result, the excessive concentration of user interest is unfavorable for excavation of potential points of interest. In this paper, we proposed a reranking user-based collaborative filtering algorithm, which generated a new recommendation list via reranking of TOP-N on the original list. The experimental results show that this algorithm can greatly improve the diversity of the final recommendation list at the sacrifice of certain accuracy. This algorithm helps users to know more previously unknown fields as well as the sales of long-tail commodities.

Paper Details