In recent years social networks have increasingly been used to study political opinion formation, monitor electoral campaigns and predict electoral outcomes as they are able to generate huge amount of data, usually in textual and non structured form. In this paper we aim at collecting and analysing data from Twitter posts identifying emerging patterns of topics related to a constitutional referendum that recently took place in Italy to better understand and nowcast its outcome. Using the Twitter API we collect tweets expressing voting intentions in the four weeks before the elections obtaining a database of approximately one million tweets. We restrict the data collection to tweets that contain hashtags referring to the referendum, therefore we are sure to include in the analysis only relevant text. On this huge volume of data, we perform a topic modelling analysis using a Latent Dirichelet Allocation model (LDA) to extract frequent topics and keywords. Analysing the behaviour of frequent words we find that connected to voting in favour of the constitutional reform there are positive words such as future and change while connected to voting against it there are words such ad fear and risk.

Using Twitter Data to Monitor Political Campaigns and Predict Election Results

FANO, SHIRA;SLANZI, Debora
2017-01-01

Abstract

In recent years social networks have increasingly been used to study political opinion formation, monitor electoral campaigns and predict electoral outcomes as they are able to generate huge amount of data, usually in textual and non structured form. In this paper we aim at collecting and analysing data from Twitter posts identifying emerging patterns of topics related to a constitutional referendum that recently took place in Italy to better understand and nowcast its outcome. Using the Twitter API we collect tweets expressing voting intentions in the four weeks before the elections obtaining a database of approximately one million tweets. We restrict the data collection to tweets that contain hashtags referring to the referendum, therefore we are sure to include in the analysis only relevant text. On this huge volume of data, we perform a topic modelling analysis using a Latent Dirichelet Allocation model (LDA) to extract frequent topics and keywords. Analysing the behaviour of frequent words we find that connected to voting in favour of the constitutional reform there are positive words such as future and change while connected to voting against it there are words such ad fear and risk.
2017
Trends in Cyber-Physical Multi-Agent Systems. The PAAMS Collection - 15th International Conference, PAAMS 2017
File in questo prodotto:
File Dimensione Formato  
2017_AISC.pdf

non disponibili

Tipologia: Documento in Post-print
Licenza: Accesso chiuso-personale
Dimensione 238.35 kB
Formato Adobe PDF
238.35 kB Adobe PDF   Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/3690238
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 3
  • ???jsp.display-item.citation.isi??? ND
social impact