Un approccio basato su NLP per analizzare l’impatto dello stile linguistico sulla popolarità degli utenti nei social media. Social networks can be considered the main protagonists of the technological revolution of the last twenty years. These platforms have in fact changed the way of knowing new people and communicating with them, giving the possibility to get in touch with anyone in every moment. At the same time, with the development of new professional figures such as influencers, the potential of these social networks was also discovered in the marketing field. In this thesis, we propose an investigation to derive an “identikit” of popular users identifying features in order to distinguish them from non-popular ones. The extraction of these features was carried out considering Twitter posts and studying not only user’s data but also the content of tweets and the quality of the language used. In the last part of this work, by using a machine learning algorithm, the information previously extracted is used for user popularity prediction. The results obtained from this approach can be considered relevant and can be used as a starting point for future studies in this research context.
An NLP-based approach to analyzing the impact of the language style on user popularity in social media
DELLA SCIUCCA, LAURA
2020/2021
Abstract
Un approccio basato su NLP per analizzare l’impatto dello stile linguistico sulla popolarità degli utenti nei social media. Social networks can be considered the main protagonists of the technological revolution of the last twenty years. These platforms have in fact changed the way of knowing new people and communicating with them, giving the possibility to get in touch with anyone in every moment. At the same time, with the development of new professional figures such as influencers, the potential of these social networks was also discovered in the marketing field. In this thesis, we propose an investigation to derive an “identikit” of popular users identifying features in order to distinguish them from non-popular ones. The extraction of these features was carried out considering Twitter posts and studying not only user’s data but also the content of tweets and the quality of the language used. In the last part of this work, by using a machine learning algorithm, the information previously extracted is used for user popularity prediction. The results obtained from this approach can be considered relevant and can be used as a starting point for future studies in this research context.È consentito all'utente scaricare e condividere i documenti disponibili a testo pieno in UNITESI UNIPV nel rispetto della licenza Creative Commons del tipo CC BY NC ND.
Per maggiori informazioni e per verifiche sull'eventuale disponibilità del file scrivere a: unitesi@unipv.it.
https://hdl.handle.net/20.500.14239/13927