Models of influence in online social networks

Kanna Alfalahi, Yacine Atif, Ajith Abraham

Research output: Contribution to journalArticlepeer-review

17 Citations (Scopus)

Abstract

Online social networks gained their popularity from relationships users can build with each other. These social ties play an important role in asserting users' behaviors in a social network. For example, a user might purchase a product that his friend recently bought. Such phenomenon is called social influence, which is used to study users' behavior when the action of one user can affect the behavior of his neighbors in a social network. Social influence is increasingly investigated nowadays as it can help spreading messages widely, particularly in the context of marketing, to rapidly promote products and services based on social friends' behavior in the network. This wide interest in social influence raises the need to develop models to evaluate the rate of social influence. In this paper, we discuss metrics used to measure influence probabilities. Then, we reveal means to maximize social influence by identifying and using the most influential users in a social network. Along with these contributions, we also survey existing social influence models, and classify them into an original categorization framework. Then, based on our proposed metrics, we show the results of an experimental evaluation to compare the influence power of some of the surveyed salient models used to maximize social influence.

Original languageEnglish
Pages (from-to)161-183
Number of pages23
JournalInternational Journal of Intelligent Systems
Volume29
Issue number2
DOIs
Publication statusPublished - Feb 2014

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Human-Computer Interaction
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Models of influence in online social networks'. Together they form a unique fingerprint.

Cite this