Mining web query logs to analyze political issues

Ingmar Weber, Venkata Rama Kiran Garimella, Erik Borra

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

29 Citations (Scopus)

Abstract

We present a novel approach to using anonymized web search query logs to analyze and visualize political issues. Our starting point is a list of politically annotated blogs (left vs. right). We use this list to assign a numerical political leaning to queries leading to clicks on these blogs. Furthermore, we map queries to Wikipedia articles and to fact-checked statements from politifact.com, as well as applying sentiment analysis to search results. With this rich, multi-faceted data set we obtain novel graphical visualizations of issues and discover connections between the different variables. Our findings include (i) an interest in "the other side" where queries about Democrat politicians have a right leaning and vice versa, (ii) evidence that "lies are catchy" and that queries pertaining to false statements are more likely to attract large volumes, and (iii) the observation that the more right-leaning a query it is, the more negative sentiments can be found in its search results.

Original languageEnglish
Title of host publicationProceedings of the 3rd Annual ACM Web Science Conference, WebSci'12
Pages330-339
Number of pages10
DOIs
Publication statusPublished - 19 Nov 2012
Externally publishedYes
Event3rd Annual ACM Web Science Conference, WebSci 2012 - Evanston, IL, United States
Duration: 22 Jun 201224 Jun 2012

Other

Other3rd Annual ACM Web Science Conference, WebSci 2012
CountryUnited States
CityEvanston, IL
Period22/6/1224/6/12

Fingerprint

Blogs
Visualization

Keywords

  • Opinion mining and sentiment analysis
  • Partisanship
  • Political leaning
  • Web search logs

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Weber, I., Garimella, V. R. K., & Borra, E. (2012). Mining web query logs to analyze political issues. In Proceedings of the 3rd Annual ACM Web Science Conference, WebSci'12 (pp. 330-339) https://doi.org/10.1145/2380718.2380761

Mining web query logs to analyze political issues. / Weber, Ingmar; Garimella, Venkata Rama Kiran; Borra, Erik.

Proceedings of the 3rd Annual ACM Web Science Conference, WebSci'12. 2012. p. 330-339.

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

Weber, I, Garimella, VRK & Borra, E 2012, Mining web query logs to analyze political issues. in Proceedings of the 3rd Annual ACM Web Science Conference, WebSci'12. pp. 330-339, 3rd Annual ACM Web Science Conference, WebSci 2012, Evanston, IL, United States, 22/6/12. https://doi.org/10.1145/2380718.2380761
Weber I, Garimella VRK, Borra E. Mining web query logs to analyze political issues. In Proceedings of the 3rd Annual ACM Web Science Conference, WebSci'12. 2012. p. 330-339 https://doi.org/10.1145/2380718.2380761
Weber, Ingmar ; Garimella, Venkata Rama Kiran ; Borra, Erik. / Mining web query logs to analyze political issues. Proceedings of the 3rd Annual ACM Web Science Conference, WebSci'12. 2012. pp. 330-339
@inproceedings{4557cdc43c3b4d6b8e6ee4df7038eee3,
title = "Mining web query logs to analyze political issues",
abstract = "We present a novel approach to using anonymized web search query logs to analyze and visualize political issues. Our starting point is a list of politically annotated blogs (left vs. right). We use this list to assign a numerical political leaning to queries leading to clicks on these blogs. Furthermore, we map queries to Wikipedia articles and to fact-checked statements from politifact.com, as well as applying sentiment analysis to search results. With this rich, multi-faceted data set we obtain novel graphical visualizations of issues and discover connections between the different variables. Our findings include (i) an interest in {"}the other side{"} where queries about Democrat politicians have a right leaning and vice versa, (ii) evidence that {"}lies are catchy{"} and that queries pertaining to false statements are more likely to attract large volumes, and (iii) the observation that the more right-leaning a query it is, the more negative sentiments can be found in its search results.",
keywords = "Opinion mining and sentiment analysis, Partisanship, Political leaning, Web search logs",
author = "Ingmar Weber and Garimella, {Venkata Rama Kiran} and Erik Borra",
year = "2012",
month = "11",
day = "19",
doi = "10.1145/2380718.2380761",
language = "English",
isbn = "9781450312288",
pages = "330--339",
booktitle = "Proceedings of the 3rd Annual ACM Web Science Conference, WebSci'12",

}

TY - GEN

T1 - Mining web query logs to analyze political issues

AU - Weber, Ingmar

AU - Garimella, Venkata Rama Kiran

AU - Borra, Erik

PY - 2012/11/19

Y1 - 2012/11/19

N2 - We present a novel approach to using anonymized web search query logs to analyze and visualize political issues. Our starting point is a list of politically annotated blogs (left vs. right). We use this list to assign a numerical political leaning to queries leading to clicks on these blogs. Furthermore, we map queries to Wikipedia articles and to fact-checked statements from politifact.com, as well as applying sentiment analysis to search results. With this rich, multi-faceted data set we obtain novel graphical visualizations of issues and discover connections between the different variables. Our findings include (i) an interest in "the other side" where queries about Democrat politicians have a right leaning and vice versa, (ii) evidence that "lies are catchy" and that queries pertaining to false statements are more likely to attract large volumes, and (iii) the observation that the more right-leaning a query it is, the more negative sentiments can be found in its search results.

AB - We present a novel approach to using anonymized web search query logs to analyze and visualize political issues. Our starting point is a list of politically annotated blogs (left vs. right). We use this list to assign a numerical political leaning to queries leading to clicks on these blogs. Furthermore, we map queries to Wikipedia articles and to fact-checked statements from politifact.com, as well as applying sentiment analysis to search results. With this rich, multi-faceted data set we obtain novel graphical visualizations of issues and discover connections between the different variables. Our findings include (i) an interest in "the other side" where queries about Democrat politicians have a right leaning and vice versa, (ii) evidence that "lies are catchy" and that queries pertaining to false statements are more likely to attract large volumes, and (iii) the observation that the more right-leaning a query it is, the more negative sentiments can be found in its search results.

KW - Opinion mining and sentiment analysis

KW - Partisanship

KW - Political leaning

KW - Web search logs

UR - http://www.scopus.com/inward/record.url?scp=84869007880&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84869007880&partnerID=8YFLogxK

U2 - 10.1145/2380718.2380761

DO - 10.1145/2380718.2380761

M3 - Conference contribution

AN - SCOPUS:84869007880

SN - 9781450312288

SP - 330

EP - 339

BT - Proceedings of the 3rd Annual ACM Web Science Conference, WebSci'12

ER -