Master's thesis

Abstract

The role of women within online communication is gaining attention. Unfortunately, it is observed a growth in hate and sexist attitudes towards them. The exposure to sexist speech is extremely harmful for both women and society. Previous studies on sexism detection have focused on identifying misogyny or hatred towards women. However, the so-called “everyday sexism” in the web has not been previously targeted. It includes subtle sexist attitudes that, although frequently unnoticed, are extremely harmful for both women and society. In the light of this, we pursue three research objectives: (i) understand how sexist behaviors, beliefs and attitudes are expressed on social networks; (ii) study the feasibility of using machine learning techniques for automatically detecting different types of sexist attitudes and expressions; (iii) inspect the spread of sexism on the Internet and analyze platforms which are promoting this type of messages.