A little bird told me your gender: Gender inferences in social media

Online and social media platforms employ automated recognition methods to presume user preferences, sensitive attributes such as race, gender, sexual orientation, opinions. These opaque can predict behaviors for marketing purposes influence behavior profit, serving attention economics but also reinforcing existing biases gender stereotyping. Although two international human rights treaties include explicit obligations relating harmful wrongful stereotyping, these stereotypes persist online offline. By identifying how inferential analytics may reinforce stereotyping affect marginalized communities, opportunities addressing concerns thereby increasing privacy, diversity, inclusion be explored. This is important because misgendering reinforces stereotypes, accentuates binarism, undermines privacy autonomy, cause feelings of rejection, impacting people's self-esteem, confidence, authenticity. In turn, this increase stigmatization. study brings into view discrimination exacerbation that continue replicate literature starts highlight. The implications on Twitter are investigated illustrate the impact algorithmic bias inadvertent violations reinforcement prejudices through a multidisciplinary perspective, including legal, computer science, critical feminist media-studies viewpoints. An pilot survey was conducted better understand accurate Twitter's inferences its users’ identities are. served basis exploring practice.