Sign prediction by conformity in signed social network

Pao Yue-kong Library Electronic Theses Database

Sign prediction by conformity in signed social network


Author: Wen, Yingting
Title: Sign prediction by conformity in signed social network
Degree: M.Sc.
Year: 2014
Subject: Online social networks
Social networks
Hong Kong Polytechnic University -- Dissertations
Department: Dept. of Computing
Pages: vii, 55 leaves : illustrations (some color) ; 30 cm
Language: English
InnoPac Record:
Abstract: Sign prediction problem in signed social network has often focused on trust/distrust propagation algorithm in a "web of trust". While in this research, we pay more attention on the discovering some underlying mechanisms of link sign formation. Conformity, as a result of try to do the same decision as the group did to fit in, is type of behavior under social influence within group context. Most of the past social psychology research only apply the theory in experiments of small groups. In our study, a machine-learning framework that include conformity and other features is used in large online social network dataset Epinions and Slashdot. The conformity feature is defined in three levels: conformity, non-conformity and counter-conformity. The model formalizes and quantifies the effect of different conformity features, which provide another approach to deal with the sign prediction problem: the group context of users when adding a positive/negative link has an influence on their decision. The model also improves the predictive accuracy to 97.17% and remains a stable performance in different datasets across domains, which indicates that it captures important principles that can be generalized.

Files in this item

Files Size Format
b27576735.pdf 1.890Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.


Quick Search


More Information