This thesis presents a user modeling pipeline to analyze discussions and opinions shared on social media regarding polarized political events (e.g., public polls). The pipeline follows a four-step methodology. First, social media posts and users metadata are crawled. Second, a filtering mechanism is applied to filter spammers and bot users. As a third step, demographics information is extracted out of the valid users, namely gender, age, ethnicity and location information. Finally, the political polarity of the users with respect to the analyzed event is predicted. In the scope of this work, our proposed pipeline is applied to three referendum scenarios: independence of Catalonia in Spain, autonomy of Lombardy in Italy and constitution amendment in Italy. Our purpose is to assess the performance of the approach with respect to the capability of collecting correct insights on the demographics of social media users and of predicting the poll results based on the opinions shared by the users. Experiments show that the method was effective in part in predicting the political trends for the Catalonia and Constitutional case, but not for the Lombardy case. Among the various motivations for this, we noticed that in general Twitter was more representative of the users opposing the referendum than the ones in favor.
Questa tesi presenta una pipeline per la modellizzazione utente per analizzare nei social media discussioni e opinioni concernenti eventi politici polarizzati (ad esempio, elezioni pubbliche). La pipeline si articola in quattro passaggi. In primo luogo, i post e i metadata degli utenti sono raccolti. In secondo luogo, un meccanismo di filtraggio viene applciato sugli spammer e utenti bot. Come terzo passaggio, le informazioni demografiche sono estratte dagli utenti validi, in particolare genere, età, etnia e luogo di residenza. Per ultimo, la polarità politica degli utenti nei confronti dell’evento analizzato viene predetto. Nel contesto del presente lavoro, la nostra pipeline viene applicata su tre scenari di referendum: indipendenza della Catalogna in Spagna, autonomia della Lombardia in Italia e emendamenti costituzionali in Italia. Il nostro scopo è quello di valutare l’efficacia dell’approccio rispetto all’abilità di trarre conclusioni corrette riguardo la demografia degli utenti nei social media e di predire i risultati delle elezioni sulla base delle opinioni espresse dagli utenti. L’esperimento ha dimostrato che il metodo è stato parzialmente efficace a predire l’orientamento politico dei casi Catalogna e Costituzionale, ma non del caso Lombardia. Tra le varie motivazioni, abbiamo rilevato che in generale in Twitter c’era stata una maggiore rappresentanza di utenti contro il referendum rispetto a quelli a favore.
A user modeling pipeline for studying polarized political events in social media
NAPOLI, ROBERTO
2017/2018
Abstract
This thesis presents a user modeling pipeline to analyze discussions and opinions shared on social media regarding polarized political events (e.g., public polls). The pipeline follows a four-step methodology. First, social media posts and users metadata are crawled. Second, a filtering mechanism is applied to filter spammers and bot users. As a third step, demographics information is extracted out of the valid users, namely gender, age, ethnicity and location information. Finally, the political polarity of the users with respect to the analyzed event is predicted. In the scope of this work, our proposed pipeline is applied to three referendum scenarios: independence of Catalonia in Spain, autonomy of Lombardy in Italy and constitution amendment in Italy. Our purpose is to assess the performance of the approach with respect to the capability of collecting correct insights on the demographics of social media users and of predicting the poll results based on the opinions shared by the users. Experiments show that the method was effective in part in predicting the political trends for the Catalonia and Constitutional case, but not for the Lombardy case. Among the various motivations for this, we noticed that in general Twitter was more representative of the users opposing the referendum than the ones in favor.File | Dimensione | Formato | |
---|---|---|---|
2018_07_napoli.pdf
non accessibile
Dimensione
1.16 MB
Formato
Adobe PDF
|
1.16 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/142105