Inflaming public debate: a methodology to determine origin and characteristics of hate speech about sexual and gender diversity on Twitter

Abstract

This article is focused on the reproduction of ideologically charged messages whose origins or interests remain hidden from public opinion. There is an urgent need for transparency regarding polarised debates that deform, impede or distort the critical approach that any society should be able to construct concerning issues of great social interest, especially on social media platforms and networks. Research has shown that hostility has colonised digital communication through misogynist, homophobic, transphobic or xenophobic messages, among others, and that, for the most part, these are not spontaneous or individual interactions. In the virtual space, there are forces that, although invisible outside it, construct narratives, generate disinformation and feed generally regressive ideological approaches. Thus, in the name of transparency and social justice, there is an urgent need to investigate these types of messages, as well as their possible destabilising interests at a time of special presence and reputation of discourses such as the feminist one, which is currently experiencing a significant reactionary response. This paper investigates the origin and characteristics of the conversation on the social network Twitter concerning gender and sexual identities. To this end, we studied a significant sample of tweets (>1 million) related to women’s rights, the LGBTIQ+ collective and trans people, for a full year. Computerised methodologies by means of machine learning techniques, natural language processing (NLP), determination of bots, geolocation, and the application of network theories were used to carry out the study. The results include the highly interrelated presence of groups without clear referents, as well as the existence of what appear to be coordinated networks aimed at causing harm and provoking confrontation.