Reproduce “More than Memes: A Multimodal Topic Modeling Approach to Conspiracy Theories on Telegram” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Elisabeth Steffen
Investigating German-language Telegram channels, this study employed multimodal topic modeling to analyze conspiracy theories across text, images, and combined text-image data. It found that while geopolitical events like the Israel-Gaza conflict dominated content, photos and screenshots, not memes, were the primary visual types, and different modalities captured distinct aspects of shared topics, leading to a conceptual framework of eight textual and visual discursive strategies.
Key insights
- Multimodal topic modeling offers more comprehensive insights into online discourse, as each modality often identifies unique document sets for shared topics, complementing insights from other modalities.
- The bulk of visual content in conspiracist communication consists of photos and screenshots, with memes playing a comparatively minor role, suggesting a broader focus is needed in visual CT research.
- Topic representation is influenced by both the input modality and the specific theme, indicating distinct communicative roles and granularities for text and visuals in conveying conspiracist narratives.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.