Kazakh Language Dataset for Hate Speech Detection on Social Media Text
Milana Bolatbek, Moldir Sagynay, Shynar Mussiraliyeva, Kymbat Baisylbayeva, Zhastay Yeltay · 2024
The article presents an urgent problem of the spread of destructive messages in the modern information society, aggravated by the rapid development of the Internet and social networks. The main attention is paid to the characterization of destructive messages, which include racism, national extremism, bullying and extremist content that can damage interpersonal relationships, self-esteem and the mental state of people. A significant increase in the number of such messages has been highlighted due to the anonymity and accessibility of Internet platforms, which leads to increased aggression and problems in online communication. The use of various methods, such as machine learning algorithms, keyword analysis, allows to effectively detect and combat destructive messages in various environments, including the Internet. This not only helps to prevent negative impacts on people, but also helps to maintain the moral and ethical standards of society. In this paper classes of destructive messages such as bullying, racism, national extremism and violent extremism are considered, emphasizing the need for an integrated approach to combating these phenomena to ensure the safety of the information space.