Abstract

Disagreements, oppositions and negative opinions are indispensable parts of online political debates. In social media, people express their beliefs and attitudes not only on issues but also about each other through both their conversations and platform-specific interactions such as like, share in Facebook and retweet in Twitter. While there are explicit "like" features in these platforms, there is no explicit "dislike" feature. Many network analysis tasks, such as detecting communities and monitoring their dynamics (i.e. polarization patterns) require information about both positive and negative linkages. Hence, predicting negative links between users is an important task and a challenging problem. In this study, we propose an unsupervised framework to predict the negative links between users by utilizing explicit positive interactions and sentiment cues in conversations. We show the effectiveness of the proposed framework on a political Twitter dataset annotated through Amazon MTurk crowdsourcing platform. Our experimental results show that the proposed framework outperforms other well-known methods and proposed baselines. To illustrate the contribution of the predicted negative links, we compare the community detection accuracies using signed and unsigned user networks. Experimental results using predicted negative links show superiority on three political datasets where the camps are known a priori. We also present qualitative evaluations related to the polarization patterns (i.e. rivalries and coalitions) between the detected communities which is only possible in the presence of negative links.

Original languageEnglish (US)
Title of host publicationHT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media
PublisherAssociation for Computing Machinery, Inc
Pages125-134
Number of pages10
ISBN (Electronic)9781450347082
DOIs
StatePublished - Jul 4 2017
Event28th ACM Conference on Hypertext and Social Media, HT 2017 - Prague, Czech Republic
Duration: Jul 4 2017Jul 7 2017

Other

Other28th ACM Conference on Hypertext and Social Media, HT 2017
CountryCzech Republic
CityPrague
Period7/4/177/7/17

Fingerprint

Polarization
Electric network analysis
Monitoring

Keywords

  • Negative Link Prediction
  • Online Political Networks
  • Sentiment Analysis
  • Social Media Mining

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Artificial Intelligence
  • Human-Computer Interaction
  • Software

Cite this

Ozer, M., Yildirim, M. Y., & Davulcu, H. (2017). Negative link prediction and its applications in online political networks. In HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media (pp. 125-134). Association for Computing Machinery, Inc. https://doi.org/10.1145/3078714.3078727

Negative link prediction and its applications in online political networks. / Ozer, Mert; Yildirim, Mehmet Yigit; Davulcu, Hasan.

HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media. Association for Computing Machinery, Inc, 2017. p. 125-134.

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

Ozer, M, Yildirim, MY & Davulcu, H 2017, Negative link prediction and its applications in online political networks. in HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media. Association for Computing Machinery, Inc, pp. 125-134, 28th ACM Conference on Hypertext and Social Media, HT 2017, Prague, Czech Republic, 7/4/17. https://doi.org/10.1145/3078714.3078727
Ozer M, Yildirim MY, Davulcu H. Negative link prediction and its applications in online political networks. In HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media. Association for Computing Machinery, Inc. 2017. p. 125-134 https://doi.org/10.1145/3078714.3078727
Ozer, Mert ; Yildirim, Mehmet Yigit ; Davulcu, Hasan. / Negative link prediction and its applications in online political networks. HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media. Association for Computing Machinery, Inc, 2017. pp. 125-134
@inproceedings{580b586d490e40d38799010e88a1982e,
title = "Negative link prediction and its applications in online political networks",
abstract = "Disagreements, oppositions and negative opinions are indispensable parts of online political debates. In social media, people express their beliefs and attitudes not only on issues but also about each other through both their conversations and platform-specific interactions such as like, share in Facebook and retweet in Twitter. While there are explicit {"}like{"} features in these platforms, there is no explicit {"}dislike{"} feature. Many network analysis tasks, such as detecting communities and monitoring their dynamics (i.e. polarization patterns) require information about both positive and negative linkages. Hence, predicting negative links between users is an important task and a challenging problem. In this study, we propose an unsupervised framework to predict the negative links between users by utilizing explicit positive interactions and sentiment cues in conversations. We show the effectiveness of the proposed framework on a political Twitter dataset annotated through Amazon MTurk crowdsourcing platform. Our experimental results show that the proposed framework outperforms other well-known methods and proposed baselines. To illustrate the contribution of the predicted negative links, we compare the community detection accuracies using signed and unsigned user networks. Experimental results using predicted negative links show superiority on three political datasets where the camps are known a priori. We also present qualitative evaluations related to the polarization patterns (i.e. rivalries and coalitions) between the detected communities which is only possible in the presence of negative links.",
keywords = "Negative Link Prediction, Online Political Networks, Sentiment Analysis, Social Media Mining",
author = "Mert Ozer and Yildirim, {Mehmet Yigit} and Hasan Davulcu",
year = "2017",
month = "7",
day = "4",
doi = "10.1145/3078714.3078727",
language = "English (US)",
pages = "125--134",
booktitle = "HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Negative link prediction and its applications in online political networks

AU - Ozer, Mert

AU - Yildirim, Mehmet Yigit

AU - Davulcu, Hasan

PY - 2017/7/4

Y1 - 2017/7/4

N2 - Disagreements, oppositions and negative opinions are indispensable parts of online political debates. In social media, people express their beliefs and attitudes not only on issues but also about each other through both their conversations and platform-specific interactions such as like, share in Facebook and retweet in Twitter. While there are explicit "like" features in these platforms, there is no explicit "dislike" feature. Many network analysis tasks, such as detecting communities and monitoring their dynamics (i.e. polarization patterns) require information about both positive and negative linkages. Hence, predicting negative links between users is an important task and a challenging problem. In this study, we propose an unsupervised framework to predict the negative links between users by utilizing explicit positive interactions and sentiment cues in conversations. We show the effectiveness of the proposed framework on a political Twitter dataset annotated through Amazon MTurk crowdsourcing platform. Our experimental results show that the proposed framework outperforms other well-known methods and proposed baselines. To illustrate the contribution of the predicted negative links, we compare the community detection accuracies using signed and unsigned user networks. Experimental results using predicted negative links show superiority on three political datasets where the camps are known a priori. We also present qualitative evaluations related to the polarization patterns (i.e. rivalries and coalitions) between the detected communities which is only possible in the presence of negative links.

AB - Disagreements, oppositions and negative opinions are indispensable parts of online political debates. In social media, people express their beliefs and attitudes not only on issues but also about each other through both their conversations and platform-specific interactions such as like, share in Facebook and retweet in Twitter. While there are explicit "like" features in these platforms, there is no explicit "dislike" feature. Many network analysis tasks, such as detecting communities and monitoring their dynamics (i.e. polarization patterns) require information about both positive and negative linkages. Hence, predicting negative links between users is an important task and a challenging problem. In this study, we propose an unsupervised framework to predict the negative links between users by utilizing explicit positive interactions and sentiment cues in conversations. We show the effectiveness of the proposed framework on a political Twitter dataset annotated through Amazon MTurk crowdsourcing platform. Our experimental results show that the proposed framework outperforms other well-known methods and proposed baselines. To illustrate the contribution of the predicted negative links, we compare the community detection accuracies using signed and unsigned user networks. Experimental results using predicted negative links show superiority on three political datasets where the camps are known a priori. We also present qualitative evaluations related to the polarization patterns (i.e. rivalries and coalitions) between the detected communities which is only possible in the presence of negative links.

KW - Negative Link Prediction

KW - Online Political Networks

KW - Sentiment Analysis

KW - Social Media Mining

UR - http://www.scopus.com/inward/record.url?scp=85026351593&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85026351593&partnerID=8YFLogxK

U2 - 10.1145/3078714.3078727

DO - 10.1145/3078714.3078727

M3 - Conference contribution

SP - 125

EP - 134

BT - HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media

PB - Association for Computing Machinery, Inc

ER -