Sentiment Analysis of Portuguese Political Parties Communication

Carlos J. Costa, Manuela Aparicio, Joao Aparicio

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

10 Citations (Scopus)
71 Downloads (Pure)

Abstract

Political communication in social media has gained increasing importance in the last years. In this study, we analyze the political parties’ communication on Twitter and understand the sentiment of their communication. First by identifying their communication performance regarding the daily number of tweets, favorite tweets, number of retweets per day and per political party. We present a sentiment analysis by the political party using tweets data. In this study, we propose an explanatory model with the main drivers of retweets. To conduct this study, our approach used data analysis and machine learning techniques methods. Results indicate the main determinants that influence future retweets of political posts globally. Here we present a comparison of the communication content between tweets posts and the political parties’ programs available on their institutional websites. We identify the similarities between tweets and formal programs per party and among all parties. This study contributes to analyze the coherence and effectiveness of the political parties’ communication.
Original languageEnglish
Title of host publicationProceedings of the 39th ACM International Conference on the Design of Communication (SIGDOC '21)
PublisherACM - Association for Computing Machinery
Pages63-69
Number of pages7
ISBN (Print)978-1-4503-8628-9
DOIs
Publication statusPublished - 12 Oct 2021
Event39th ACM International Conference on the Design of Communication: Building Coalitions. Worldwide, SIGDOC 2021 - Virtual, Online, United States
Duration: 12 Oct 202114 Oct 2021
Conference number: 39

Conference

Conference39th ACM International Conference on the Design of Communication: Building Coalitions. Worldwide, SIGDOC 2021
Abbreviated titleSIGDOC 2021
Country/TerritoryUnited States
Period12/10/2114/10/21

Keywords

  • document similarity
  • machine learning
  • political parties
  • sentiment analysis
  • Twitter

Fingerprint

Dive into the research topics of 'Sentiment Analysis of Portuguese Political Parties Communication'. Together they form a unique fingerprint.

Cite this