Telegram Post
Telegram, a widely used messaging app, is the subject of increasing research focusing on understanding and mitigating its use for the spread of misinformation, propaganda, and harmful content. Current studies employ various machine learning techniques, including transformer-based models like BERT and LLMs such as GPT-4, to detect conspiracy theories, human rights violations, and pro-Kremlin propaganda across multiple languages. This research is crucial for developing tools to combat disinformation and improve online safety, with implications for content moderation, fact-checking, and the broader understanding of information manipulation in online social networks.
Papers
Partial Mobilization: Tracking Multilingual Information Flows Amongst Russian Media Outlets and Telegram
Hans W. A. Hanley, Zakir Durumeric
Automated multilingual detection of Pro-Kremlin propaganda in newspapers and Telegram posts
Veronika Solopova, Oana-Iuliana Popescu, Christoph Benzmüller, Tim Landgraf