Perception of western leaders by anonymous ukrainian telegram news channels
Keywords:
artificial intelligence, sentiment analysis, political communication, the full-scale Russian invasion of Ukraine, social networks, information content, public communications, digitalization, digital technologiesAbstract
The article is devoted to sentiment analysis of texts about Western political leaders by anonymous Ukrainian Telegram news channels in the period from the beginning of the full-scale Russian invasion of Ukraine to the end of 2024. The purpose of the study is to assess the relevance of using the Ukrainian library based on artificial intelligence (Ukrainian Roberta base) for automatic determination of sentiments in texts. The material for the study was exported publications of the most popular anonymous Telegram channels "Trukha" and "Ukraine Now" for the specified period, from which fragments containing mentions of Western leaders (Joe Biden, Donald Trump, Olaf Scholz, Boris Johnson, Rishi Sunak, Emmanuel Macron) were selected. The methodology included quantitative counting of mentions and application of sentiment analysis to classify them (positive, negative, neutral). The analysis was carried out using the Ukrainian Roberta base model, which estimates the probability of a text belonging to a certain sentiment. The results showed a predominance of neutral mentions in both channels, but Trukha Ukraine turned out to be more emotional, with a higher percentage of negative and very negative sentiments compared to Ukraine Now. The most negative mentions concerned Donald Trump and Olaf Scholz (for Ukraine Now) and Emmanuel Macron (for Trukha). At the same time, the most positive sentiments were recorded for Boris Johnson (Ukraine Now), Rishi Sunak and Joe Biden (Trukha). An important observation was that negative sentiment often arose due to the presence of negative particles or words with a negative connotation, and not due to an actual negative attitude of the media. Positive sentiments were mostly formed due to quotes from politicians that contained words of support and emojis. That is, the presence of positive or negative sentiments does not mean a corresponding attitude of the media. The conclusions emphasize that AI-based sentiment analysis is relevant for large datasets, but its results should be interpreted with caution, as they reflect linguistic markers, not the position of the media. The study also revealed the Ukrainian-centricity of the content of Telegram channels: mentions of Western leaders are mostly associated with their support for Ukraine or rejection of it. The results obtained may be useful for further research into the informational impact of anonymous Telegram channels and assessing the effectiveness of automatic sentiment analysis tools in the Ukrainian media space.
References
Споживання новин і довіра до медіа в Україні. URL: https://internews.ua/media-research
Стеблина Н. Нам пишуть, що щось коїться – як регіональні телеграм-канали інформують українців під час Великої війни. URL: https://idpo.org.ua/reports/6136-nam-pishut-shho-shhos-ko%d1%97tsya-yak-regionalni-telegram-kanali-informuyut-ukra%d1%97nciv-pid-chas-veliko%d1%97-vijni.html
Баркар Д. Телеграм-канали як інструмент. Хто і як формував підтримку закону № 12414 та атакував НАБУ – дослідження ІМІ. URL: https://imi.org.ua/monitorings/telegram-kanaly-yak-instrument-hto-i-yak-formuvav-pidtrymku-zakonu-12414-ta-atakuvav-nabu-i69667
Медіавійна проти НАБУ і САП. Як топові телеграм-канали намагаються знищити репутацію антикорупційних інституцій. URL: https://detector.media/infospace/article/245610/2025-11-14-mediaviyna-proty-nabu-i-sap-yak-topovi-telegram-kanaly-namagayutsya-znyshchyty-reputatsiyu-antykoruptsiynykh-instytutsiy/
Як телеграм-канали заробляють на страхах українців: основні маніпуляції. URL: https://texty.org.ua/fragments/115168/yak-telehram-kanaly-zaroblyayut-na-strahah-ukrayinciv-osnovni-manipulyaciyi/
Машкова Я. Джинса і маніпуляції. Як заробляють українські телеграм-канали з мільйонними аудиторіями? URL: https://detector.media/infospace/article/218087/2023-10-14-dzhynsa-i-manipulyatsii-yak-zaroblyayut-ukrainski-telegram-kanaly-z-milyonnymy-audytoriyamy/
Власник Telegram-каналу "Труха" Лавриненко: Один дуже-дуже відомий олігарх сказав: "Хочу купити ваш канал зі збереженням команди". Пропонував $5 млн. URL: https://gordonua.com/ukr/news/money/vlasnik-telegram-kanalu-trukha-lavrinenko-odin-duzhe-duzhe-vidomij-oliharkh-skazav-khochu-kupiti-vash-kanal-zi-zberezhennjam-komandi-proponuvav-5-mln-1687779.html
Неплій А. “Нас читають російські воєнні блогери”. Інтерв’ю з автором Telegram-каналу “Лачен пише”. URL: https://www.kyivpost.com/uk/post/12105
Коджушко Я. Великі гравці не мають бути анонімними: власник "України на часі" Вадим Інгеров пояснив, як треба регулювати телеграм-канали в Україні. URL: https://informator.ua/uk/veliki-gravci-ne-mayut-buti-anonimnimi-vlasnik-ukrajini-na-chasi-vadim-ingerov-poyasniv-yak-treba-regulyuvati-telegram-kanali-v-ukrajini
Kermanidis, K.L., & Maragoudakis, M. (2013). Political sentiment analysis of tweets before and after the Greek elections of May 2012. Int. J. Soc. Netw. Min., 1, 298-317.
Matalon, Y., Magdaci, O., Almozlino, A., & Yamin, D. (2021). Using sentiment analysis to predict opinion inversion in Tweets of political communication. Scientific Reports, 11.
Tumasjan, Andranik & Sprenger, Timm & Sandner, Philipp & Welpe, Isabell. (2010). Predicting Elections with Twitter: What 140 Characters Reveal about Political Sentiment. Word. Journal Of The International Linguistic Association. 10.
Stieglitz, S., & Dang-Xuan, L. (2013). Emotions and Information Diffusion in Social Media—Sentiment of Microblogs and Sharing Behavior. Journal of Management Information Systems, 29, 217 - 248.
Antypas, D., Preece, A.D., & Camacho-Collados, J. (2022). Negativity spreads faster: A large-scale multilingual twitter analysis on the role of sentiment in political communication. Online Soc. Networks Media, 33, 100242.
Ott, B. L. (2017). The age of Twitter: Donald J. Trump and the politics of debasement. Critical Studies in Media Communication, 34(1), 59–68. https://doi.org/10.1080/15295036.2016.1266686
Методологія оцінювання якості контенту в регіональних друкованих та он-лайн виданнях. URL: https://idpo.org.ua/wp-content/uploads/methodology-2023.pdf
Haselmayer, M., & Jenny, M. (2016). Sentiment analysis of political communication: combining a dictionary approach with crowdcoding. Quality & Quantity, 51, 2623 - 2646.
Стеблина Н. Цифрові інструменти для медіа-аналітика. Одеса: Фенікс, 2023.
Yang, X., Zhu, J.Y., & Meo, P.D. (2024). A quantum-like zero-shot approach for sentiment analysis in finance. Journal of Intelligent Information Systems, 63, 705 - 721.
