| Beschreibung |
Social media has become a significant part of our every day life, and continues to change the way we interact with each other on a global scale. In this project, we work on the development of computational tools and methods for the analysis of social media at large scale. A core task will be the development of a neural representation for social media feeds (i.e., sequences of posts that may be commented and liked) using pre-trained multimodal neural networks from Hugging Face (https://huggingface.co/). The envisioned representation is supposed to capture the topical preferences of a feed, including topic-specific stances. Along the work on representations, we will work on an analysis toolbox facilitating among others the clustering of feeds into user groups, the classification of feeds, e.g. with respect to specific stances, and even the generation of feed-related posts. |
