Misinformation on Twitter During the Danish National Election: A Case Study

Leon Derczynski; Torben Oskar Albert-Lindqvist; Marius Venø Bendsen; Nanna Inie; Jens Egholm Pedersen; Viktor Due Pedersen

Misinformation on Twitter During the Danish National Election: A Case Study

Leon Derczynski
, Torben Oskar Albert-Lindqvist
, Marius Venø Bendsen
, Nanna Inie
, Jens Egholm Pedersen
, Viktor Due Pedersen

Lix Technologies ApS

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

Elections are a time when communication is important in democracies, including over social media. This paper describes a case study of applying NLP to determine the extent to which misinformation and external manipulation were present on Twitter during a national election. We use three methods to detect the spread of misinformation: analysing unusual spatial and temporal behaviours; detecting known false claims and using these to estimate the total prevalence; and detecting amplifiers through language use. We find that while present, detectable spread of misinformation on Twitter was remarkably low during the election period in Denmark.

Original language	English
Title of host publication	Proceedings of the conference on Truth and Trust Online
Publication date	4 Oct 2019
Publication status	Published - 4 Oct 2019
Event	Truth and Trust Online - London, United Kingdom Duration: 4 Oct 2019 → 5 Oct 2019 Conference number: 1 https://truthandtrustonline.com

Conference

Conference	Truth and Trust Online
Number	1
Country/Territory	United Kingdom
City	London
Period	04/10/2019 → 05/10/2019
Internet address	https://truthandtrustonline.com

Keywords

Elections
Social Media
NLP
Misinformation Detection
Twitter Analysis

Access to Document

paper_16Final published version, 965 KB

https://truthandtrustonline.files.wordpress.com/2019/09/paper_16.pdf

Cite this

@inproceedings{2f158a6e31564a0fa28ca55b1556f1d0,

title = "Misinformation on Twitter During the Danish National Election: A Case Study",

abstract = "Elections are a time when communication is important in democracies, including over social media. This paper describes a case study of applying NLP to determine the extent to which misinformation and external manipulation were present on Twitter during a national election. We use three methods to detect the spread of misinformation: analysing unusual spatial and temporal behaviours; detecting known false claims and using these to estimate the total prevalence; and detecting amplifiers through language use. We find that while present, detectable spread of misinformation on Twitter was remarkably low during the election period in Denmark.",

keywords = "Elections, Social Media, NLP, Misinformation Detection, Twitter Analysis, Elections, Social Media, NLP, Misinformation Detection, Twitter Analysis",

author = "Leon Derczynski and Albert-Lindqvist, \{Torben Oskar\} and Bendsen, \{Marius Ven{\o}\} and Nanna Inie and Pedersen, \{Jens Egholm\} and Pedersen, \{Viktor Due\}",

note = "The proceedings are published online only, without ISBN or DOI, and published by the conference directly by uploading. This is a new conference and not BFI indexed. ; Truth and Trust Online, TTO ; Conference date: 04-10-2019 Through 05-10-2019",

year = "2019",

month = oct,

day = "4",

language = "English",

booktitle = "Proceedings of the conference on Truth and Trust Online",

url = "https://truthandtrustonline.com",

}

TY - GEN

T1 - Misinformation on Twitter During the Danish National Election: A Case Study

AU - Derczynski, Leon

AU - Albert-Lindqvist, Torben Oskar

AU - Bendsen, Marius Venø

AU - Inie, Nanna

AU - Pedersen, Jens Egholm

AU - Pedersen, Viktor Due

N1 - Conference code: 1

PY - 2019/10/4

Y1 - 2019/10/4

N2 - Elections are a time when communication is important in democracies, including over social media. This paper describes a case study of applying NLP to determine the extent to which misinformation and external manipulation were present on Twitter during a national election. We use three methods to detect the spread of misinformation: analysing unusual spatial and temporal behaviours; detecting known false claims and using these to estimate the total prevalence; and detecting amplifiers through language use. We find that while present, detectable spread of misinformation on Twitter was remarkably low during the election period in Denmark.

AB - Elections are a time when communication is important in democracies, including over social media. This paper describes a case study of applying NLP to determine the extent to which misinformation and external manipulation were present on Twitter during a national election. We use three methods to detect the spread of misinformation: analysing unusual spatial and temporal behaviours; detecting known false claims and using these to estimate the total prevalence; and detecting amplifiers through language use. We find that while present, detectable spread of misinformation on Twitter was remarkably low during the election period in Denmark.

KW - Elections

KW - Social Media

KW - NLP

KW - Misinformation Detection

KW - Twitter Analysis

KW - Elections

KW - Social Media

KW - NLP

KW - Misinformation Detection

KW - Twitter Analysis

M3 - Article in proceedings

BT - Proceedings of the conference on Truth and Trust Online

T2 - Truth and Trust Online

Y2 - 4 October 2019 through 5 October 2019

ER -

Misinformation on Twitter During the Danish National Election: A Case Study

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this