|Title:||Persuasion-driven social influence analysis in social networks|
|Advisors:||Chung, Korris Fu-lai (COMP)|
|Subject:||Online social networks.|
Digital communications -- Social aspects.
Hong Kong Polytechnic University -- Dissertations
|Department:||Department of Computing|
|Pages:||x, 71 leaves : illustrations ; 30 cm|
|Abstract:||It is universally believed that people are increasingly relying on online social network to work, study, and communicate with others nowadays. Thus, it is unavoidable for us to be influenced by others as in real world, more or less. This has attracted many researchers' attention to study information diffusion and social influence in social networks. Among them, many works have been done to solve the influence maximization problem. For instance, Tang et al. proposed a famous Topical Affinity Propagation (TAP) model to analyze the social influence on topic level. The TAP model manages to identify the most influential nodes on a given topic in social networks successfully. It contributes to the solution of one fundamental question in influence maximization problem: How possible are we influenced? Aside from this, Leung et al. presents a Persuasiveness Aware Cascade (PAC) model to quantitatively estimate the peer influence probability from a social persuasion perspective in sociology, and further study the information propagation problem. It succeeds in addressing another fundamental question: How can we maximize the influence spread? Actually, TAP model mainly focuses on the topic factor and considers little about the complex relationship between individuals, which is important in calculating the influence probabilities between nodes. In PAC model, there is a significant improvement in describing real-world influence propagation with the new social persuasion measures. Motivated by the above works, in order to analyze the social influence more comprehensively, we utilize the topical information and the characteristics of individuals' relationship by employing two types of social persuasiveness principles, in social networks together in our new proposed model, namely, PacTAP model. Finally, experiments conducted on three types of real-world social networks suggest that the new proposed PacTAP model outperforms the TAP model overwhelmingly in identifying influential nodes in accuracy by 2% to 36% on average, while not in efficiency.|
|Rights:||All rights reserved|
Files in This Item:
|b2780026x.pdf||For All Users (off-campus access for PolyU Staff & Students only)||2.85 MB||Adobe PDF||View/Open|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- 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.
- 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.
Please use this identifier to cite or link to this item: