Identifying political topics in social media messages: A lexicon-based approach

Sam Jackson; Feifei Zhang; Olga Boichak; Lauren Bryant; Yingya Li; Jeff Hemsley; Jennifer Stromer-Galley; Bryan Semaan; Nancy McCracken

doi:10.1145/3097286.3097298

Identifying political topics in social media messages: A lexicon-based approach

Sam Jackson, Feifei Zhang, Olga Boichak, Lauren Bryant, Yingya Li, Jeff Hemsley, Jennifer Stromer-Galley, Bryan Semaan, Nancy McCracken

School of Information Studies

Research output: Chapter in Book/Entry/Poem › Conference contribution

7 Scopus citations

Abstract

In this paper, we introduce a lexicon-based method for identifying political topics in social media messages. After discussing several critical shortcomings of unsupervised topic identification for this task, we describe the lexicon-based approach. We test our lexicon on candidate-generated campaign messages on Facebook and Twitter in the 2016 U.S. presidential election. The results show that this approach provides reliable results for eight of nine political topic categories. In closing, we describe steps to improve our approach and how it can be used for future research on political topics in social media messages.

Original language	English (US)
Title of host publication	8th International Conference on Social Media and Society
Subtitle of host publication	Social Media for Good or Evil, #SMSociety 2017
Publisher	Association for Computing Machinery
ISBN (Electronic)	9781450348478
DOIs	https://doi.org/10.1145/3097286.3097298
State	Published - Jul 28 2017
Event	8th International International Conference on Social Media and Society, #SMSociety 2017 - Toronto, Canada Duration: Jul 28 2017 → Jul 30 2017

Publication series

Name	ACM International Conference Proceeding Series
Volume	Part F129683

Other

Other	8th International International Conference on Social Media and Society, #SMSociety 2017
Country/Territory	Canada
City	Toronto
Period	7/28/17 → 7/30/17

Keywords

Document classification
Political issues
Presidential campaigns
Social media
political communication
topic identification

ASJC Scopus subject areas

Software
Human-Computer Interaction
Computer Vision and Pattern Recognition
Computer Networks and Communications

Access to Document

10.1145/3097286.3097298

Cite this

Jackson, S., Zhang, F., Boichak, O., Bryant, L., Li, Y., Hemsley, J., Stromer-Galley, J., Semaan, B., & McCracken, N. (2017). Identifying political topics in social media messages: A lexicon-based approach. In 8th International Conference on Social Media and Society: Social Media for Good or Evil, #SMSociety 2017 Article 3097298 (ACM International Conference Proceeding Series; Vol. Part F129683). Association for Computing Machinery. https://doi.org/10.1145/3097286.3097298

Identifying political topics in social media messages: A lexicon-based approach. / Jackson, Sam; Zhang, Feifei; Boichak, Olga et al.
8th International Conference on Social Media and Society: Social Media for Good or Evil, #SMSociety 2017. Association for Computing Machinery, 2017. 3097298 (ACM International Conference Proceeding Series; Vol. Part F129683).

Research output: Chapter in Book/Entry/Poem › Conference contribution

Jackson, S, Zhang, F, Boichak, O, Bryant, L, Li, Y, Hemsley, J , Stromer-Galley, J , Semaan, B & McCracken, N 2017, Identifying political topics in social media messages: A lexicon-based approach. in 8th International Conference on Social Media and Society: Social Media for Good or Evil, #SMSociety 2017., 3097298, ACM International Conference Proceeding Series, vol. Part F129683, Association for Computing Machinery, 8th International International Conference on Social Media and Society, #SMSociety 2017, Toronto, Canada, 7/28/17. https://doi.org/10.1145/3097286.3097298

Jackson S, Zhang F, Boichak O, Bryant L, Li Y, Hemsley J et al. Identifying political topics in social media messages: A lexicon-based approach. In 8th International Conference on Social Media and Society: Social Media for Good or Evil, #SMSociety 2017. Association for Computing Machinery. 2017. 3097298. (ACM International Conference Proceeding Series). doi: 10.1145/3097286.3097298

@inproceedings{8273083728ba4a01830016dd5d489f8b,

title = "Identifying political topics in social media messages: A lexicon-based approach",

abstract = "In this paper, we introduce a lexicon-based method for identifying political topics in social media messages. After discussing several critical shortcomings of unsupervised topic identification for this task, we describe the lexicon-based approach. We test our lexicon on candidate-generated campaign messages on Facebook and Twitter in the 2016 U.S. presidential election. The results show that this approach provides reliable results for eight of nine political topic categories. In closing, we describe steps to improve our approach and how it can be used for future research on political topics in social media messages.",

keywords = "Document classification, Political issues, Presidential campaigns, Social media, political communication, topic identification",

author = "Sam Jackson and Feifei Zhang and Olga Boichak and Lauren Bryant and Yingya Li and Jeff Hemsley and Jennifer Stromer-Galley and Bryan Semaan and Nancy McCracken",

note = "Publisher Copyright: {\textcopyright} 2017 Copyright is held by the owner/author(s). Publication rights licensed to ACM.; 8th International International Conference on Social Media and Society, #SMSociety 2017 ; Conference date: 28-07-2017 Through 30-07-2017",

year = "2017",

month = jul,

day = "28",

doi = "10.1145/3097286.3097298",

language = "English (US)",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

booktitle = "8th International Conference on Social Media and Society",

}

TY - GEN

T1 - Identifying political topics in social media messages

T2 - 8th International International Conference on Social Media and Society, #SMSociety 2017

AU - Jackson, Sam

AU - Zhang, Feifei

AU - Boichak, Olga

AU - Bryant, Lauren

AU - Li, Yingya

AU - Hemsley, Jeff

AU - Stromer-Galley, Jennifer

AU - Semaan, Bryan

AU - McCracken, Nancy

PY - 2017/7/28

Y1 - 2017/7/28

N2 - In this paper, we introduce a lexicon-based method for identifying political topics in social media messages. After discussing several critical shortcomings of unsupervised topic identification for this task, we describe the lexicon-based approach. We test our lexicon on candidate-generated campaign messages on Facebook and Twitter in the 2016 U.S. presidential election. The results show that this approach provides reliable results for eight of nine political topic categories. In closing, we describe steps to improve our approach and how it can be used for future research on political topics in social media messages.

AB - In this paper, we introduce a lexicon-based method for identifying political topics in social media messages. After discussing several critical shortcomings of unsupervised topic identification for this task, we describe the lexicon-based approach. We test our lexicon on candidate-generated campaign messages on Facebook and Twitter in the 2016 U.S. presidential election. The results show that this approach provides reliable results for eight of nine political topic categories. In closing, we describe steps to improve our approach and how it can be used for future research on political topics in social media messages.

KW - Document classification

KW - Political issues

KW - Presidential campaigns

KW - Social media

KW - political communication

KW - topic identification

UR - http://www.scopus.com/inward/record.url?scp=85028690316&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028690316&partnerID=8YFLogxK

U2 - 10.1145/3097286.3097298

DO - 10.1145/3097286.3097298

M3 - Conference contribution

AN - SCOPUS:85028690316

T3 - ACM International Conference Proceeding Series

BT - 8th International Conference on Social Media and Society

PB - Association for Computing Machinery

Y2 - 28 July 2017 through 30 July 2017

ER -

Identifying political topics in social media messages: A lexicon-based approach

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this