On social media platforms, and Twitter in particular, specific classes of users such as influencers have been given satisfactory operational definitions in terms of network and content metrics. Others, for instance online activists, are not less important but their characterisation still requires experimenting. We make the hypothesis that such interesting users can be found within temporally and spatially localised contexts, i.e., small but topical fragments of the network containing interactions about social events or campaigns with a significant footprint on Twitter. To explore this hypothesis, we have designed a continuous user profile discovery pipeline that produces an ever-growing dataset of user profiles by harvesting and analysing contexts from the Twitter stream. The profiles dataset includes key network topology and content-based users metrics, enabling experimentation with user-defined score functions that characterise specific classes of online users. The paper describes the design and implementation of the pipeline and its empirical evaluation on a case study consisting of healthcare-related campaigns in the UK, showing how it supports the operational definitions of online activism, by comparing three experimental ranking functions. This project is part of a joint program of the Newcastle University (UK) and the Federal University of Rio Grande Do Norte (Brazil) in an effort to fight mosquito-borne diseases such as Zika, Dengue and Chikungunya. This project has been presented and published (doi:10.1007/978-3-030-19274-7_8) at the 2019 edition of the ICWE conference in South Korea (ICWE International Conference on Web Engineering - 19th edition - South Korea, Daejeon - https://icwe2019.webengineering.org). The source code is publicly available (https://github.com/flaprimo/twitter-network-analysis).
Alcune categorie di utenti come gli influencers, nei social media, e in Twitter in particolare, hanno definizioni operative soddisfacenti in termini di metriche di topologia di rete e contenuti. Altre, ad esempio gli attivisti online, non sono meno importanti ma la loro caratterizzazione richiede un approfondimento. Ipotizziamo che questa interessante categoria di utenti possa essere trovata in contesti localizzati a livello temporale e spaziale, ovvero, piccoli frammenti topici di reti contenenti interazioni relative ad eventi o campagne social con una significativa presenza su Twitter. Per esplorare questa ipotesi abbiamo progettato una procedura per la scoperta continua di profili di utenti, che produce un dataset di profili in continua evoluzione, raccogliendo e analizzando contesti da Twitter. Il dataset include importanti metriche riguardanti la topologia di rete e i contenuti pubblicati dagli utenti, permettendo la sperimentazione di funzioni personalizzate di ordinamento dei profili degli utenti online, caratterizzandoli in specifiche classi di appartenenza. Il paper descrive la progettazione e l'implementazione della procedura e la sua valutazione empirica su un caso di studio riguardante le campagne social della sanità britannica, mostrando come supporta la definizione operativa dell'attivismo online, nella comparazione di tre funzioni di ordinamento sperimentali. Questo progetto fa parte di un programma di lavoro congiunto tra l'Università di Newcastle (Regno Unito) e l'Università Federale del Rio Grande Do Norte (Brasile) per combattere le malattie trasmesse dalle zanzare, come Zika, Dengue e Chikunguya. Questo progetto è stato presentato e pubblicato (doi:10.1007/978-3-030-19274-7_8) durante l'edizione del 2019 della conferenza ICWE (ICWE International Conference on Web Engineering - 19th edition - South Korea, Daejeon - https://icwe2019.webengineering.org) nella Corea del Sud. Il codice sorgente è pubblicamente disponibile (http:s//github.com/flaprimo/twitter-network-analysis).
A customisable pipeline for continuously harvesting socially-minded Twitter users
PRIMO, FLAVIO
2019/2020
Abstract
On social media platforms, and Twitter in particular, specific classes of users such as influencers have been given satisfactory operational definitions in terms of network and content metrics. Others, for instance online activists, are not less important but their characterisation still requires experimenting. We make the hypothesis that such interesting users can be found within temporally and spatially localised contexts, i.e., small but topical fragments of the network containing interactions about social events or campaigns with a significant footprint on Twitter. To explore this hypothesis, we have designed a continuous user profile discovery pipeline that produces an ever-growing dataset of user profiles by harvesting and analysing contexts from the Twitter stream. The profiles dataset includes key network topology and content-based users metrics, enabling experimentation with user-defined score functions that characterise specific classes of online users. The paper describes the design and implementation of the pipeline and its empirical evaluation on a case study consisting of healthcare-related campaigns in the UK, showing how it supports the operational definitions of online activism, by comparing three experimental ranking functions. This project is part of a joint program of the Newcastle University (UK) and the Federal University of Rio Grande Do Norte (Brazil) in an effort to fight mosquito-borne diseases such as Zika, Dengue and Chikungunya. This project has been presented and published (doi:10.1007/978-3-030-19274-7_8) at the 2019 edition of the ICWE conference in South Korea (ICWE International Conference on Web Engineering - 19th edition - South Korea, Daejeon - https://icwe2019.webengineering.org). The source code is publicly available (https://github.com/flaprimo/twitter-network-analysis).| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi_Magistrale.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
2.65 MB
Formato
Adobe PDF
|
2.65 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/10589/165300