A novel recommender system fusing the opinions from experts and ordinary people
Pei Jung Wu, Weiping Liu, Ci-Hang Jin · 2010
In this paper, we propose a novel recommendation algorithm fusing the opinions from experts and ordinary people. Instead of regarding one's judgement capability as his/her expertise, we present a new definition which measures the amount of the recommendable items one know in a certain area. When computing the expertise, we consider both the average value and the accumulative value, and introduce a free parameter α to tune between these two values. To evaluate the proposed algorithm, simulations are run on the Moviepilot dataset, and the results demonstrate that our algorithm outperforms the conventional collaborative filtering algorithm.