Alleviating bias leads to accurate and personalized recommendation

¹ School of Information Engineering, Nanchang Hangkong University - Nanchang 330063, PRC
² Institute of Information Economy, Hangzhou Normal University - Hangzhou 310036, PRC
³ Web Sciences Center, University of Electronic Science and Technology of China - Chengdu 610054, PRC
⁴ School of Finance and Coordinated Innovation Center of Wealth Management and Quantitative Investment, Zhejiang University of Finance and Economics - Hangzhou 310018, PRC

^(a) tianqiu.edu@gmail.com
^(b) zhangzike@gmail.com

Received: 27 September 2013
Accepted: 25 November 2013

Abstract

Recommendation bias towards objects has been found to have an impact on personalized recommendation, since objects present heterogeneous characteristics in some network-based recommender systems. In this article, based on a biased heat conduction recommendation algorithm (BHC) which considers the heterogeneity of the target objects, we propose a heterogeneous heat conduction algorithm (HHC), by further taking the heterogeneity of the source objects into account. Tested on three real datasets, the Netflix, RYM and MovieLens, the HHC algorithm is found to present better recommendation in both the accuracy and diversity than two benchmark algorithms, i.e., the original BHC and a hybrid algorithm of heat conduction and mass diffusion (HHM), while not requiring any other accessorial information or parameter. Moreover, the HHC algorithm also elevates the recommendation accuracy on cold objects, referring to the so-called cold-start problem. Eigenvalue analyses show that, the HHC algorithm effectively alleviates the recommendation bias towards objects with different level of popularity, which is beneficial to solving the accuracy-diversity dilemma.