An End-to-End Framework to Identify Pathogenic Social Media Accounts on Twitter

Elham Shaabani, Ashkan Sadeghi Mobarakeh, Hamidreza Alvari, Paulo Shakarian

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Pathogenic Social Media (PSM) accounts such as terrorist supporter accounts and fake news writers have the capability of spreading disinformation to viral proportions. Early detection of PSM accounts is crucial as they are likely to be key users to make malicious information 'viral'. In this paper, we adopt the causal inference framework along with graph-based metrics in order to distinguish PSMs from normal users within a short time of their activities. We propose both supervised and semi-supervised approaches without taking the network information and content into account. Results on a real-world the dataset from Twitter accentuates the advantage of our proposed frameworks. We show our approach achieves 0.28 improvement in F1 score over existing approaches with the precision of 0.90 and F1 score of 0.63.

Original languageEnglish (US)
Title of host publicationProceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages128-135
Number of pages8
ISBN (Electronic)9781728120805
DOIs
StatePublished - Jun 2019
Externally publishedYes
Event2nd International Conference on Data Intelligence and Security, ICDIS 2019 - South Padre Island, United States
Duration: Jun 28 2019Jun 30 2019

Publication series

NameProceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019

Conference

Conference2nd International Conference on Data Intelligence and Security, ICDIS 2019
CountryUnited States
CitySouth Padre Island
Period6/28/196/30/19

Keywords

  • bipartite graph analysis
  • causal reasoning
  • deep learning approach
  • Fake account detection
  • Pathogenic account

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Hardware and Architecture
  • Safety, Risk, Reliability and Quality

Cite this

Shaabani, E., Sadeghi Mobarakeh, A., Alvari, H., & Shakarian, P. (2019). An End-to-End Framework to Identify Pathogenic Social Media Accounts on Twitter. In Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019 (pp. 128-135). [8855274] (Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICDIS.2019.00027

An End-to-End Framework to Identify Pathogenic Social Media Accounts on Twitter. / Shaabani, Elham; Sadeghi Mobarakeh, Ashkan; Alvari, Hamidreza; Shakarian, Paulo.

Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 128-135 8855274 (Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Shaabani, E, Sadeghi Mobarakeh, A, Alvari, H & Shakarian, P 2019, An End-to-End Framework to Identify Pathogenic Social Media Accounts on Twitter. in Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019., 8855274, Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019, Institute of Electrical and Electronics Engineers Inc., pp. 128-135, 2nd International Conference on Data Intelligence and Security, ICDIS 2019, South Padre Island, United States, 6/28/19. https://doi.org/10.1109/ICDIS.2019.00027
Shaabani E, Sadeghi Mobarakeh A, Alvari H, Shakarian P. An End-to-End Framework to Identify Pathogenic Social Media Accounts on Twitter. In Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 128-135. 8855274. (Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019). https://doi.org/10.1109/ICDIS.2019.00027
Shaabani, Elham ; Sadeghi Mobarakeh, Ashkan ; Alvari, Hamidreza ; Shakarian, Paulo. / An End-to-End Framework to Identify Pathogenic Social Media Accounts on Twitter. Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 128-135 (Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019).
@inproceedings{6006117567484112b9c96d2d5bd40229,
title = "An End-to-End Framework to Identify Pathogenic Social Media Accounts on Twitter",
abstract = "Pathogenic Social Media (PSM) accounts such as terrorist supporter accounts and fake news writers have the capability of spreading disinformation to viral proportions. Early detection of PSM accounts is crucial as they are likely to be key users to make malicious information 'viral'. In this paper, we adopt the causal inference framework along with graph-based metrics in order to distinguish PSMs from normal users within a short time of their activities. We propose both supervised and semi-supervised approaches without taking the network information and content into account. Results on a real-world the dataset from Twitter accentuates the advantage of our proposed frameworks. We show our approach achieves 0.28 improvement in F1 score over existing approaches with the precision of 0.90 and F1 score of 0.63.",
keywords = "bipartite graph analysis, causal reasoning, deep learning approach, Fake account detection, Pathogenic account",
author = "Elham Shaabani and {Sadeghi Mobarakeh}, Ashkan and Hamidreza Alvari and Paulo Shakarian",
year = "2019",
month = "6",
doi = "10.1109/ICDIS.2019.00027",
language = "English (US)",
series = "Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "128--135",
booktitle = "Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019",

}

TY - GEN

T1 - An End-to-End Framework to Identify Pathogenic Social Media Accounts on Twitter

AU - Shaabani, Elham

AU - Sadeghi Mobarakeh, Ashkan

AU - Alvari, Hamidreza

AU - Shakarian, Paulo

PY - 2019/6

Y1 - 2019/6

N2 - Pathogenic Social Media (PSM) accounts such as terrorist supporter accounts and fake news writers have the capability of spreading disinformation to viral proportions. Early detection of PSM accounts is crucial as they are likely to be key users to make malicious information 'viral'. In this paper, we adopt the causal inference framework along with graph-based metrics in order to distinguish PSMs from normal users within a short time of their activities. We propose both supervised and semi-supervised approaches without taking the network information and content into account. Results on a real-world the dataset from Twitter accentuates the advantage of our proposed frameworks. We show our approach achieves 0.28 improvement in F1 score over existing approaches with the precision of 0.90 and F1 score of 0.63.

AB - Pathogenic Social Media (PSM) accounts such as terrorist supporter accounts and fake news writers have the capability of spreading disinformation to viral proportions. Early detection of PSM accounts is crucial as they are likely to be key users to make malicious information 'viral'. In this paper, we adopt the causal inference framework along with graph-based metrics in order to distinguish PSMs from normal users within a short time of their activities. We propose both supervised and semi-supervised approaches without taking the network information and content into account. Results on a real-world the dataset from Twitter accentuates the advantage of our proposed frameworks. We show our approach achieves 0.28 improvement in F1 score over existing approaches with the precision of 0.90 and F1 score of 0.63.

KW - bipartite graph analysis

KW - causal reasoning

KW - deep learning approach

KW - Fake account detection

KW - Pathogenic account

UR - http://www.scopus.com/inward/record.url?scp=85066905216&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066905216&partnerID=8YFLogxK

U2 - 10.1109/ICDIS.2019.00027

DO - 10.1109/ICDIS.2019.00027

M3 - Conference contribution

AN - SCOPUS:85066905216

T3 - Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019

SP - 128

EP - 135

BT - Proceedings - 2019 2nd International Conference on Data Intelligence and Security, ICDIS 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -