People feel the need to learn about what is happening around them, keep themselves updated and nurture their curiosity. This has never changed. The means by which to satisfy this desire of news instead, evolve together with history and technology. We believe that in the near future we will witness a rapid change of hierarchies in news-media sources and that the dissemination of news will be increasingly based on Social Media interactions. In recent years new forms of communication are emerging and the role traditionally assigned to newspapers and journalists is intertwining with unseen actors on the news-spreaders stage. Studying people’s behavior on social networks can help shed light on these changes and perspectives on the information world. This work deals with the problem of representing users in relation to their involvement in the news-spreading process. We propose an enriched user model that supports traditional content-based profiling techniques with an additional dimension built relying on both explicit and latent features of newspaper articles. We have also designed a process of analysis able to generate a large collection of users, represented according to the aforementioned model, and which only requires a list of newspapers as input.
Le persone sentono il bisogno di conoscere ciò che accade intorno a loro, si tengono aggiornati e alimentano la propria curiosità. Questo non è mai cambiato. I mezzi con cui soddisfare questo desiderio di sapere, invece, si evolvono insieme alla storia e alla tecnologia. Crediamo che nel prossimo futuro assisteremo a un rapido cambiamento nelle gerarchie delle fonti di informazione e che la diffusione delle notizie sarà sempre più basata sulle interazioni nei Social Media. Negli ultimi anni stanno emergendo nuove forme di comuni- cazione e il tradizionale ruolo di quotidiani e giornalisti si intreccia con nuove figure che partecipano al processo di informazione. Studiare il comportamento delle persone sui social network può aiutare a far luce su questi cambiamenti e sulle prospettive future per mondo dell’informazione. Questo lavoro affronta il problema di rappresentare gli utenti in relazione al loro coinvol- gimento nel processo di diffusione delle notizie. Proponiamo un modello utente arricchito, che supporta le tradizionali tecniche di profiling basate sui contenuti con una dimensione ag- giuntiva che coinvolge le caratteristiche esplicite e latenti degli articoli di giornale. Abbiamo inoltre progettato un processo di analisi in grado di generare un’ampia collezione di utenti, rappresentata secondo il suddetto modello, e che richiede come input solanto un elenco di quotidiani.
Modeling and profiling of users based on news-sharing behavior on social media
BRENA, GIOVANNI
2017/2018
Abstract
People feel the need to learn about what is happening around them, keep themselves updated and nurture their curiosity. This has never changed. The means by which to satisfy this desire of news instead, evolve together with history and technology. We believe that in the near future we will witness a rapid change of hierarchies in news-media sources and that the dissemination of news will be increasingly based on Social Media interactions. In recent years new forms of communication are emerging and the role traditionally assigned to newspapers and journalists is intertwining with unseen actors on the news-spreaders stage. Studying people’s behavior on social networks can help shed light on these changes and perspectives on the information world. This work deals with the problem of representing users in relation to their involvement in the news-spreading process. We propose an enriched user model that supports traditional content-based profiling techniques with an additional dimension built relying on both explicit and latent features of newspaper articles. We have also designed a process of analysis able to generate a large collection of users, represented according to the aforementioned model, and which only requires a list of newspapers as input.File | Dimensione | Formato | |
---|---|---|---|
2018_12_Brena.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
3.55 MB
Formato
Adobe PDF
|
3.55 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/10589/144781