Scalable sentiment classification across multiple dark web forums

David Zimbra; Hsinchun Chen

doi:10.1109/ISI.2012.6284095

Scalable sentiment classification across multiple dark web forums

David Zimbra, Hsinchun Chen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

9 Scopus citations

Abstract

This study examines several approaches to sentiment classification in the Dark Web Forum Portal, and opportunities to transfer classifiers and text features across multiple forums to improve scalability and performance. Although sentiment classifiers typically perform poorly when transferred across domains, experimentation reveals the devised approaches offer performance equivalent to the traditional forum-specific approach in classification in an unknown domain. Furthermore, incorporating the text features identified as significant indicators of sentiment in other forums can greatly improve the classification accuracy of the traditional forum-specific approach.

Original language	English (US)
Title of host publication	ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics
Subtitle of host publication	Cyberspace, Border, and Immigration Securities
Pages	78-83
Number of pages	6
DOIs	https://doi.org/10.1109/ISI.2012.6284095
State	Published - 2012
Event	2012 10th IEEE International Conference on Intelligence and Security Informatics, ISI 2012 - Washington, DC, United States Duration: Jun 11 2012 → Jun 14 2012

Publication series

Name	ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities

Other

Other	2012 10th IEEE International Conference on Intelligence and Security Informatics, ISI 2012
Country/Territory	United States
City	Washington, DC
Period	6/11/12 → 6/14/12

Keywords

dark web
domain transfer
sentiment analysis

ASJC Scopus subject areas

Artificial Intelligence
Information Systems

Access to Document

10.1109/ISI.2012.6284095

Cite this

Zimbra, D., & Chen, H. (2012). Scalable sentiment classification across multiple dark web forums. In ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities (pp. 78-83). Article 6284095 (ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities). https://doi.org/10.1109/ISI.2012.6284095

Scalable sentiment classification across multiple dark web forums. / Zimbra, David; Chen, Hsinchun.
ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities. 2012. p. 78-83 6284095 (ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Zimbra, D & Chen, H 2012, Scalable sentiment classification across multiple dark web forums. in ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities., 6284095, ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities, pp. 78-83, 2012 10th IEEE International Conference on Intelligence and Security Informatics, ISI 2012, Washington, DC, United States, 6/11/12. https://doi.org/10.1109/ISI.2012.6284095

Zimbra D, Chen H. Scalable sentiment classification across multiple dark web forums. In ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities. 2012. p. 78-83. 6284095. (ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities). doi: 10.1109/ISI.2012.6284095

Zimbra, David ; Chen, Hsinchun. / Scalable sentiment classification across multiple dark web forums. ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities. 2012. pp. 78-83 (ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities).

@inproceedings{c0110fe8712e420e93e2e117a97a77a2,

title = "Scalable sentiment classification across multiple dark web forums",

abstract = "This study examines several approaches to sentiment classification in the Dark Web Forum Portal, and opportunities to transfer classifiers and text features across multiple forums to improve scalability and performance. Although sentiment classifiers typically perform poorly when transferred across domains, experimentation reveals the devised approaches offer performance equivalent to the traditional forum-specific approach in classification in an unknown domain. Furthermore, incorporating the text features identified as significant indicators of sentiment in other forums can greatly improve the classification accuracy of the traditional forum-specific approach.",

keywords = "dark web, domain transfer, sentiment analysis",

author = "David Zimbra and Hsinchun Chen",

year = "2012",

doi = "10.1109/ISI.2012.6284095",

language = "English (US)",

isbn = "9781467321037",

series = "ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities",

pages = "78--83",

booktitle = "ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics",

note = "2012 10th IEEE International Conference on Intelligence and Security Informatics, ISI 2012 ; Conference date: 11-06-2012 Through 14-06-2012",

}

TY - GEN

T1 - Scalable sentiment classification across multiple dark web forums

AU - Zimbra, David

AU - Chen, Hsinchun

PY - 2012

Y1 - 2012

N2 - This study examines several approaches to sentiment classification in the Dark Web Forum Portal, and opportunities to transfer classifiers and text features across multiple forums to improve scalability and performance. Although sentiment classifiers typically perform poorly when transferred across domains, experimentation reveals the devised approaches offer performance equivalent to the traditional forum-specific approach in classification in an unknown domain. Furthermore, incorporating the text features identified as significant indicators of sentiment in other forums can greatly improve the classification accuracy of the traditional forum-specific approach.

AB - This study examines several approaches to sentiment classification in the Dark Web Forum Portal, and opportunities to transfer classifiers and text features across multiple forums to improve scalability and performance. Although sentiment classifiers typically perform poorly when transferred across domains, experimentation reveals the devised approaches offer performance equivalent to the traditional forum-specific approach in classification in an unknown domain. Furthermore, incorporating the text features identified as significant indicators of sentiment in other forums can greatly improve the classification accuracy of the traditional forum-specific approach.

KW - dark web

KW - domain transfer

KW - sentiment analysis

UR - http://www.scopus.com/inward/record.url?scp=84867386556&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867386556&partnerID=8YFLogxK

U2 - 10.1109/ISI.2012.6284095

DO - 10.1109/ISI.2012.6284095

M3 - Conference contribution

AN - SCOPUS:84867386556

SN - 9781467321037

T3 - ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics: Cyberspace, Border, and Immigration Securities

SP - 78

EP - 83

BT - ISI 2012 - 2012 IEEE International Conference on Intelligence and Security Informatics

T2 - 2012 10th IEEE International Conference on Intelligence and Security Informatics, ISI 2012

Y2 - 11 June 2012 through 14 June 2012

ER -

Scalable sentiment classification across multiple dark web forums

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this