A Linguistic Analysis of Sexism-Related Hate Speech in Social Media

The aim of this article is to present the functioning of a dual algorithm/human analysis and to investigate the means with which to study hate speech, especially sexism-related hate speech, in the online environment, focusing on social media comments and hashtags. Another aim is to investigate new linguistic trends in contemporary online hate speech that can be revealed via quantitative hate speech analysis. In the first part, the concept of hate speech is briefly introduced in a linguistic context. In the second part, an example of a Twitter hashtag is analyzed. In the third part, an algorithm for the identification of sexism-related hate speech from the corpus available at hatespeechdata.com is discussed. The article demonstrates the methods of evaluating selected types of online content for the presence of hate speech. It is made evident that algorithm-based hate speech qualification is an insufficient tool for identifying hate speech and that qualitative analysis by a trained linguist is necessary.