Discovering social spammers from multiple views

Hua Shen, Fenglong Ma, Xianchao Zhang, Linlin Zong, Xinyue Liu, Wenxin Liang

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Online social networks have become popular platforms for spammers to spread malicious content and links. Existing state-of-the-art optimization methods mainly use one kind of user-generated information (i.e., single view) to learn a classification model for identifying spammers. Due to the diversity and variability of spammers' strategies, spammers' behavior may not be completely characterized only by single view's information. To tackle this challenge, we first statistically analyze the importance of considering multiple view information for spammer detection task on a large real-world Twitter dataset. Accordingly, we propose a generalized social spammer detection framework by jointly integrating multiple view information and a novel social regularization term into a classification model. To keep the completeness of the original dataset and detect more spammers by the proposed method, we introduce a simple strategy to fill the missing data for each view. Experimental results on a real-world Twitter dataset show that the proposed method outperforms the existing methods significantly.

Original languageEnglish (US)
Pages (from-to)49-57
Number of pages9
JournalNeurocomputing
Volume225
DOIs
StatePublished - Feb 15 2017

Fingerprint

Social Support
Datasets

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Shen, H., Ma, F., Zhang, X., Zong, L., Liu, X., & Liang, W. (2017). Discovering social spammers from multiple views. Neurocomputing, 225, 49-57. https://doi.org/10.1016/j.neucom.2016.11.013
Shen, Hua ; Ma, Fenglong ; Zhang, Xianchao ; Zong, Linlin ; Liu, Xinyue ; Liang, Wenxin. / Discovering social spammers from multiple views. In: Neurocomputing. 2017 ; Vol. 225. pp. 49-57.
@article{4b57cd2a408d478d91b66f80d7bec6a3,
title = "Discovering social spammers from multiple views",
abstract = "Online social networks have become popular platforms for spammers to spread malicious content and links. Existing state-of-the-art optimization methods mainly use one kind of user-generated information (i.e., single view) to learn a classification model for identifying spammers. Due to the diversity and variability of spammers' strategies, spammers' behavior may not be completely characterized only by single view's information. To tackle this challenge, we first statistically analyze the importance of considering multiple view information for spammer detection task on a large real-world Twitter dataset. Accordingly, we propose a generalized social spammer detection framework by jointly integrating multiple view information and a novel social regularization term into a classification model. To keep the completeness of the original dataset and detect more spammers by the proposed method, we introduce a simple strategy to fill the missing data for each view. Experimental results on a real-world Twitter dataset show that the proposed method outperforms the existing methods significantly.",
author = "Hua Shen and Fenglong Ma and Xianchao Zhang and Linlin Zong and Xinyue Liu and Wenxin Liang",
year = "2017",
month = "2",
day = "15",
doi = "10.1016/j.neucom.2016.11.013",
language = "English (US)",
volume = "225",
pages = "49--57",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

Shen, H, Ma, F, Zhang, X, Zong, L, Liu, X & Liang, W 2017, 'Discovering social spammers from multiple views', Neurocomputing, vol. 225, pp. 49-57. https://doi.org/10.1016/j.neucom.2016.11.013

Discovering social spammers from multiple views. / Shen, Hua; Ma, Fenglong; Zhang, Xianchao; Zong, Linlin; Liu, Xinyue; Liang, Wenxin.

In: Neurocomputing, Vol. 225, 15.02.2017, p. 49-57.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Discovering social spammers from multiple views

AU - Shen, Hua

AU - Ma, Fenglong

AU - Zhang, Xianchao

AU - Zong, Linlin

AU - Liu, Xinyue

AU - Liang, Wenxin

PY - 2017/2/15

Y1 - 2017/2/15

N2 - Online social networks have become popular platforms for spammers to spread malicious content and links. Existing state-of-the-art optimization methods mainly use one kind of user-generated information (i.e., single view) to learn a classification model for identifying spammers. Due to the diversity and variability of spammers' strategies, spammers' behavior may not be completely characterized only by single view's information. To tackle this challenge, we first statistically analyze the importance of considering multiple view information for spammer detection task on a large real-world Twitter dataset. Accordingly, we propose a generalized social spammer detection framework by jointly integrating multiple view information and a novel social regularization term into a classification model. To keep the completeness of the original dataset and detect more spammers by the proposed method, we introduce a simple strategy to fill the missing data for each view. Experimental results on a real-world Twitter dataset show that the proposed method outperforms the existing methods significantly.

AB - Online social networks have become popular platforms for spammers to spread malicious content and links. Existing state-of-the-art optimization methods mainly use one kind of user-generated information (i.e., single view) to learn a classification model for identifying spammers. Due to the diversity and variability of spammers' strategies, spammers' behavior may not be completely characterized only by single view's information. To tackle this challenge, we first statistically analyze the importance of considering multiple view information for spammer detection task on a large real-world Twitter dataset. Accordingly, we propose a generalized social spammer detection framework by jointly integrating multiple view information and a novel social regularization term into a classification model. To keep the completeness of the original dataset and detect more spammers by the proposed method, we introduce a simple strategy to fill the missing data for each view. Experimental results on a real-world Twitter dataset show that the proposed method outperforms the existing methods significantly.

UR - http://www.scopus.com/inward/record.url?scp=85006830352&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006830352&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2016.11.013

DO - 10.1016/j.neucom.2016.11.013

M3 - Article

AN - SCOPUS:85006830352

VL - 225

SP - 49

EP - 57

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -