Two voices, one war: NLP and LDA for sociological analysis of political communication on Twitter vis-a-vis the Ukrainian war
摘要
This article applies the framework of Computational Communication Science (CCS) (Theocharis and Jungherr, in Political Communication, Routledge, Milton Park 2021) to the analysis of political communication by British and Polish parliamentarians in response to the war in Ukraine. Focusing on posts published on the X platform (formerly Twitter) during the first year of the war, the study uses sentiment analysis (BERT) and topic modeling (LDA) to identify patterns in how parliamentarians from two allied countries expressed themselves in the context of a shared international crisis. While both countries consistently supported Ukraine, the analysis revealed notable differences in emotional tone: British members of parliament expressed significantly more positive sentiment overall, whereas Polish parliamentarians, despite addressing the Ukrainian topic more frequently, communicated in a more emotionally restrained and neutral manner. These patterns are described and contextualized, with some possible interpretations proposed, though the study does not aim to offer tested causal explanations. Instead, it contributes to comparative research on political discourse by demonstrating how computational tools can uncover differences in issue salience and affective expression across national contexts.