查看文章

Background: the Russian aggression against Ukraine has notably intensified online discourse on this subject. This surge in online activity inevitably results in the spread of content, some of which may be unreliable or manipulative. The identification of such content with information distortion is crucial to mitigate bias and promote trustworthiness.

Objectives: improve the efficiency of distinguishing between stance/attitude, and determining the sentiment of messages related to such events as the Russian war in Ukraine.

Importance: from a practical perspective, our piece of research can significantly contribute to cognitive security and cyber hygiene purposes. In terms of Ukrainian NLP, it would be an advancement to accurately classify geopolitical attitudes and sentiment in text data with puzzling language usage and linked to complex topics.