Detecting Dutch political tweets: A classifier based on voting system using supervised learning

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

Abstract

The task of classifying political tweets has been shown to be very difficult, with controversial results in many works and with non-replicable methods. Most of the works with this goal use rule-based methods to identify political tweets. We propose here two methods, being one rule-based approach, which has an accuracy of 62%, and a supervised learning approach, which went up to 97% of accuracy in the task of distinguishing political and non-political tweets in a corpus of 2.881 Dutch tweets. Here we show that for a data base of Dutch tweets, we can outperform the rule-based method by combining many different supervised learning methods.

LanguageEnglish
Title of host publicationICAART 2018 - Proceedings of the 10th International Conference on Agents and Artificial Intelligence
Place of PublicationSetúbal
PublisherSciTePress
Pages462-469
Number of pages8
Volume2
ISBN (Electronic)9789897582752
StatePublished - Jan 2018
Event10th International Conference on Agents and Artificial Intelligence, ICAART 2018 - Funchal, Madeira, Portugal
Duration: 16 Jan 201818 Jan 2018

Conference

Conference10th International Conference on Agents and Artificial Intelligence, ICAART 2018
CountryPortugal
CityFunchal, Madeira
Period16/01/1818/01/18

Fingerprint

Supervised learning
Classifiers

Keywords

  • Dutch politics
  • Machine Learning
  • Natural Language Processing
  • Politics
  • Text Mining
  • Twitter

Cite this

de Mello Araújo, E. F., & Ebbelaar, D. (2018). Detecting Dutch political tweets: A classifier based on voting system using supervised learning. In ICAART 2018 - Proceedings of the 10th International Conference on Agents and Artificial Intelligence (Vol. 2, pp. 462-469). Setúbal: SciTePress.
de Mello Araújo, Eric Fernandes ; Ebbelaar, Dave. / Detecting Dutch political tweets : A classifier based on voting system using supervised learning. ICAART 2018 - Proceedings of the 10th International Conference on Agents and Artificial Intelligence. Vol. 2 Setúbal : SciTePress, 2018. pp. 462-469
@inproceedings{215ca4f014734d50b6b532c5b740f8f9,
title = "Detecting Dutch political tweets: A classifier based on voting system using supervised learning",
abstract = "The task of classifying political tweets has been shown to be very difficult, with controversial results in many works and with non-replicable methods. Most of the works with this goal use rule-based methods to identify political tweets. We propose here two methods, being one rule-based approach, which has an accuracy of 62{\%}, and a supervised learning approach, which went up to 97{\%} of accuracy in the task of distinguishing political and non-political tweets in a corpus of 2.881 Dutch tweets. Here we show that for a data base of Dutch tweets, we can outperform the rule-based method by combining many different supervised learning methods.",
keywords = "Dutch politics, Machine Learning, Natural Language Processing, Politics, Text Mining, Twitter",
author = "{de Mello Ara{\'u}jo}, {Eric Fernandes} and Dave Ebbelaar",
year = "2018",
month = "1",
language = "English",
volume = "2",
pages = "462--469",
booktitle = "ICAART 2018 - Proceedings of the 10th International Conference on Agents and Artificial Intelligence",
publisher = "SciTePress",

}

de Mello Araújo, EF & Ebbelaar, D 2018, Detecting Dutch political tweets: A classifier based on voting system using supervised learning. in ICAART 2018 - Proceedings of the 10th International Conference on Agents and Artificial Intelligence. vol. 2, SciTePress, Setúbal, pp. 462-469, 10th International Conference on Agents and Artificial Intelligence, ICAART 2018, Funchal, Madeira, Portugal, 16/01/18.

Detecting Dutch political tweets : A classifier based on voting system using supervised learning. / de Mello Araújo, Eric Fernandes; Ebbelaar, Dave.

ICAART 2018 - Proceedings of the 10th International Conference on Agents and Artificial Intelligence. Vol. 2 Setúbal : SciTePress, 2018. p. 462-469.

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

TY - GEN

T1 - Detecting Dutch political tweets

T2 - A classifier based on voting system using supervised learning

AU - de Mello Araújo,Eric Fernandes

AU - Ebbelaar,Dave

PY - 2018/1

Y1 - 2018/1

N2 - The task of classifying political tweets has been shown to be very difficult, with controversial results in many works and with non-replicable methods. Most of the works with this goal use rule-based methods to identify political tweets. We propose here two methods, being one rule-based approach, which has an accuracy of 62%, and a supervised learning approach, which went up to 97% of accuracy in the task of distinguishing political and non-political tweets in a corpus of 2.881 Dutch tweets. Here we show that for a data base of Dutch tweets, we can outperform the rule-based method by combining many different supervised learning methods.

AB - The task of classifying political tweets has been shown to be very difficult, with controversial results in many works and with non-replicable methods. Most of the works with this goal use rule-based methods to identify political tweets. We propose here two methods, being one rule-based approach, which has an accuracy of 62%, and a supervised learning approach, which went up to 97% of accuracy in the task of distinguishing political and non-political tweets in a corpus of 2.881 Dutch tweets. Here we show that for a data base of Dutch tweets, we can outperform the rule-based method by combining many different supervised learning methods.

KW - Dutch politics

KW - Machine Learning

KW - Natural Language Processing

KW - Politics

KW - Text Mining

KW - Twitter

UR - http://www.scopus.com/inward/record.url?scp=85046638233&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046638233&partnerID=8YFLogxK

M3 - Conference contribution

VL - 2

SP - 462

EP - 469

BT - ICAART 2018 - Proceedings of the 10th International Conference on Agents and Artificial Intelligence

PB - SciTePress

CY - Setúbal

ER -

de Mello Araújo EF, Ebbelaar D. Detecting Dutch political tweets: A classifier based on voting system using supervised learning. In ICAART 2018 - Proceedings of the 10th International Conference on Agents and Artificial Intelligence. Vol. 2. Setúbal: SciTePress. 2018. p. 462-469.