Web User Trust Relationship Prediction Based on Cloud Model

  • Authors:
  • Qian Zhang;Tianyi Zhang;Xianglin Zuo;Yuan Yao;Baoping Feng;Junhua Wang;Wanli Zuo

  • Affiliations:
  • College of Computer Science and Technology, Jilin University 2699 Qianjin Street, Changchun 130012 People's Republic of China 86-431-85168892;College of Computer Science and Technology, Jilin University 2699 Qianjin Street, Changchun 130012 People's Republic of China 86-431-85168892;College of Computer Science and Technology, Jilin University 2699 Qianjin Street, Changchun 130012 People's Republic of China 86-431-85168892;College of Computer Science and Technology, Jilin University 2699 Qianjin Street, Changchun 130012 People's Republic of China 86-431-85168892;College of Computer Science and Technology, Jilin University 2699 Qianjin Street, Changchun 130012 People's Republic of China 86-431-85168892;College of Computer Science and Technology, Jilin University 2699 Qianjin Street, Changchun 130012 People's Republic of China 86-431-85168892;College of Computer Science and Technology, Jilin University 2699 Qianjin Street, Changchun 130012 People's Republic of China 86-431-85168892

  • Venue:
  • Proceedings of the Second International Conference on Innovative Computing and Cloud Computing
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Web user trustiness is closely related to web information credibility, which has attracted wide spread concern since the emergence of Web 2.0. Motivated by the randomness and fuzziness of user trust relationships at the web age, this paper proposes a novel user trust relationship prediction method based on cloud model. We start by defining trust cloud, distrust cloud and scoring cloud; then calculate expectation, entropy and hyper-entropy to extract digital characteristics of cloud by utilizing inverse cloud generation algorithm; and finally predict user trust relationships based on trust distance and distrust distance. To analyze the effect of the proposed approach, we conducted experiment on the Extended Epinions dataset, which shows that the precision, recall and F-measure of our method reaches 96.94%, 99.14% and 98.03% respectively, with an increase of 0.21%, 13.11%, and 9.01% over current best method, demonstrating that predicting user trust relations in social networks based on cloud model is reasonable and effective.