The Method of Personalized Recommendation with Ensemble Combination

Jiwan Seo, Seungjin Choi, Mucheol Kim, and Sangyong Han⁺

Chung-Ang University, Seoul, Republic of Korea
{jwseo, bethemoney}@ec.cse.cau.ac.kr, mckim@kisti.re.kr, hansy@cau.ac.kr

Abstract
Nowadays trust and reputation models are becoming more and more important to make the decision through various industries. Trust-based system is vulnerable to sparse relations, so there are attempts to combine the trust and reputation models. In this paper, we propose a method in which the trust and reputation models are harmonized through an ensemble combination. It can be applied to not only the personalized recommendations but also the detection of malicious insider users which attack with unfair rating. The proposed method enables both the models to complement each other and provides the reliably personalized service.

Keywords: Trust and reputation model, Recommendation system, Ensemble combination

+: Corresponding author: Sangyong Han
Department of Computer Science and Engineering, Chung-Ang University, Heukseok 1(il)-dong,
Dongjak-gu, Seoul, Republic of Korea, Tel: +82-(0)28205327

Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable Applications (JoWUA),

Vol. 4, No. 4, pp. 108-121, December 2013 [pdf]