Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Rumor blocking through online link deletion on social networks
Yan R., Li Y., Wu W., Li D., Wang Y. ACM Transactions on Knowledge Discovery from Data13 (2):1-26,2019.Type:Article
Date Reviewed: Aug 21 2019

Today, people often depend on social networking services to get information. It is often observed that rumors spread quickly over social networks, owing to personal relationships. Social networking service managers must block malicious rumors in order to keep the social networking service trustworthy. This paper proposes a rumor spread minimization (RSM) problem, which is a timely topic.

There are two basic social network diffusion models, the independent cascade (IC) model and the linear threshold (LT) model. The IC model is a simple model in which diffusion occurs depending on an independent probability, while the LT model uses a threshold of the number of activated neighbor nodes to determine a new diffusion. This paper uses an IC model to present the proposed algorithm. The proposed algorithm finds an edge set that minimizes the rumor spread value when the edge set is deleted from the social network. The rumor spread value is the objective function, defined as the sum of the probability that a node is activated by the rumor seed set. Since the proposed algorithm uses a greedy approach, the submodularity of the objective function is important to provide a performance guarantee. This paper shows that the objective function is not submodular; it proposes a submodular upper-bound and a submodular lower-bound, which can be used to express the original objective function as the difference between the two submodular functions.

The experimental results show that the proposed algorithm outperforms existing greedy algorithms, including the bond percolation method (BPM) and K edge deletion (KED), as well as heuristic algorithms such as Out-degree (OD), PageRank (PR), and Random (RAN). In several real social networks (Wikipedia, Slashdot, Google+, HEP-TH), the proposed algorithm presents the lowest rumor spread value among the existing algorithms. The results are promising, and follow-up studies for other real social networks that can be represented by the LT model are expected.

Reviewer:  Seon Yeong Han Review #: CR146664 (1911-0390)
Bookmark and Share
 
Network Architecture And Design (C.2.1 )
 
 
General (F.0 )
 
Would you recommend this review?
yes
no
Other reviews under "Network Architecture And Design": Date
Designing data networks
Ellis R., Prentice-Hall, Inc., Upper Saddle River, NJ, 1986. Type: Book (9789780132018647)
Apr 1 1986
Internetworking
Pouzin L., Prentice-Hall, Inc., Upper Saddle River, NJ, 1986. Type: Book (9789780131650503)
Feb 1 1987
Broadband network technology: an overview for the data and telecommunications industries
Cooper E., Prentice-Hall, Inc., Upper Saddle River, NJ, 1986. Type: Book (9789780130833792)
Oct 1 1987
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy