Using deep-learning algorithms to derive basic characteristics of social media users: The Brexit campaign as a case study

Abstract

A recurrent criticism concerning the use of online social media data in political science research is the lack of demographic information about social media users. By employing a face-recognition algorithm to the profile pictures of Facebook users, the paper derives two fundamental demographic characteristics (age and gender) of a sample of Facebook users who interacted with the most relevant British parties in the two weeks before the Brexit referendum of 23 June 2016. The article achieves the goals of (i) testing the precision of the algorithm, (ii) testing its validity, (iii) inferring new evidence on digital mobilisation, and (iv) tracing the path for future developments and application of the algorithm. The findings show that the algorithm is reliable and that it can be fruitfully used in political and social sciences both to confirm the validity of survey data and to obtain information from populations that are generally unavailable within traditional surveys.

Redes e sociabilidades na Modernidade Tardia

“…São sistemas que servem para "fisgar", neste caso, pessoas: "uma tendência entre os fabricantes desses sistemas de descrever o seu propósito como 'fisgar' [hook] pessoas -aliciando-as para uso frequente ou duradouro". 33 Consultar, sobre o assunto: Del Vicario (2017), Mancosu (2019). 34 Cf.…”

Modeling activity spaces using big geo‐data: Progress and challenges

“…LBSM data usually cause fewer privacy concerns than georeferenced mobile phone data and can be obtained through application program interfaces (APIs). LBSM users can also publish more detailed background information such as their age, gender, education, and employment on their public profile (Fohringer et al., 2015; Mancosu & Bobba, 2019; Yuan et al., 2018). The geolocations in LBSM can either be coordinates acquired from built‐in smart phone GPS modules if the user enabled accurate positioning or approximated POIs with a bounding box (e.g., “The City of Austin” or “Yellowstone National Park”) (Yuan et al., 2020).…”

Disinformation and Narratives

“…Social media, like in any election in modern democracies, played a significant role in the EU referendum in the UK, and there has been extensive research into the impact in particular Twitter had on the referendum and Brexit (e.g. Gorodnichenko, Pham, & Talavera, 2018;Grčar, Cherepnalkoski, Mozetič, & Kralj Novak, 2017;Hänska & Bauchowitz, 2017;Llewellyn, Cram, Hill, & Favero, 2019;Mancosu & Bobba, 2019). Researchers found that "the pro-Brexit camp was four times more influential" on Twitter than the pro-Remain side (Grčar et al, 2017, p. 1), that the predominance of Euroscepticism on Twitter mirrored the predominance of Euroscepticism in the UK press coverage of the referendum (Hänska & Bauchowitz, 2017), or that the Leave side had significant impact through spreading false information on social media (Yan, 2019).…”