Automated multilingual detection of Pro-Kremlin propaganda in newspapers and Telegram posts

TL;DR

This paper develops and compares two multilingual automated methods using Transformers and linguistic features to detect pro-Kremlin propaganda in news and social media during the Russia-Ukraine conflict, addressing misinformation challenges.

Contribution

It introduces and evaluates two novel approaches for multilingual propaganda detection, analyzing their effectiveness, adaptability, and ethical implications in a conflict context.

Findings

Transformers-based method shows high accuracy in propaganda detection.

Linguistic features offer better interpretability and adaptability.

Both methods face challenges with new genres and languages.

Abstract

The full-scale conflict between the Russian Federation and Ukraine generated an unprecedented amount of news articles and social media data reflecting opposing ideologies and narratives. These polarized campaigns have led to mutual accusations of misinformation and fake news, shaping an atmosphere of confusion and mistrust for readers worldwide. This study analyses how the media affected and mirrored public opinion during the first month of the war using news articles and Telegram news channels in Ukrainian, Russian, Romanian and English. We propose and compare two methods of multilingual automated pro-Kremlin propaganda identification, based on Transformers and linguistic features. We analyse the advantages and disadvantages of both methods, their adaptability to new genres and languages, and ethical considerations of their usage for content moderation. With this work, we aim to lay the foundation for further development of moderation tools tailored to the current conflict.