Type Like a Man! Inferring Gender from Keystroke Dynamics in Live-Chats

Citations Over TimeTop 10% of 2019 papers

Abstract

Nonverbal communication is often referred to as body language, an expression that accounts for the major role that the body plays in interaction, especially when it comes to conveying socially and psychologically relevant information. Such a role is the result of a long evolutionary process that has shaped the brain to be sensitive to the signals sent by co -located others more than to any other signal in the environment (e.g., the human voice is one of the sounds that requires the lowest energy to be heard). Still, despite such an evolutionary history, people communicate increasingly more frequently through technologies that prevent, partially or totally, the use of nonverbal behavior. For example, phones allow one to use nonverbal vocal behavior (laughter, sobbing, intonation, pauses, etc.), but not facial expressions or gestures. In the context outlined above, it is important to investigate whether body language is still possible when the body cannot play its role. For this reason, this article has shown that there is a significant interplay between gender and keystroke dynamics at least in the case of interactions taking place through live -chat interfaces. In particular, the experiments have shown that it is possible to infer the gender of a person from her typing behavior with an accuracy higher than 95%. In addition, the experiments have shown that such a performance relies mostly on features (physical and machine detectable measures extracted through a key -logging platform) that account for implicit and explicit expression of affect, social presence, and planning problems. According to a recent survey, 36% of adults owning a smartphone use messaging systems (https://www.pewinternet.org/2015/08/19/mobilemessaging-and-social-media-2015D. In addition, The market for technologies supporting live -chats is expected to 40 50 60 grow with an average rate of 7.3% until 2023 when it is expected to reach a total volume close to one billion dollars per year (https://

Related Papers

• → Examining a Large Keystroke Biometrics Dataset for Statistical-Attack Openings(2013)60 cited
• → Keystroke Dynamics as Part of Lifelogging(2021)5 cited
• → Keystroke-based authentication by key press intervals as a complementary behavioral biometric(2009)21 cited
• → Authenticating User's Keystroke Based on Statistical Models(2010)12 cited
• → A Comprehensive Analysis of Keystroke Recognition System(2021)1 cited