Abusive Speech Detection Method for Ukrainian Language Used Recurrent Neural Network
Iurii V. Krak, ОЛЬГА ЗАЛУЦЬКА, Maryna O. Molchanova, Olexander V. Mazurets, Ruslan Bahrii, Олена Собко, Olexander Barmak · 2024
The paper is devoted to the creation and approbation of a method for determining the level of social acceptability of textual Ukrainian-language content, which will be able to determine the level of detection of offensive speech for the Ukrainian language using a recurrent neural network based on the entered textual information.The method of detecting offensive speech involves the use of a combined approach based on the use of a dictionary of offensive words and a neural network approach to determine the sentiment tone of the message.The proposed method will allow to assess the level of social acceptability of Ukrainian-language Internet content for automated moderation of Internet content.Further research can be focused on applied use, which can be a useful tool for assessing the level of social acceptability of digital text content published on social networks and for preventing the spread of harmful or offensive information.It can also help improve the quality of communication and increase the level of social interaction in general.