BERTopic vs CATMA

BERTopic has no write-up on Specifiction yet; its figures are above.

CATMA is a free tool for computer-assisted text markup and analysis, designed for qualitative text studies, especially literary studies. Users can upload corpora, share projects, create tagsets, and apply overlapping annotations. Work can be exploratory or taxonomy-based, qualitative or quantitative, and individual or collaborative. Its Analyze module provides interactive visualizations such as word clouds, distribution graphs, and Double Tree. The web application runs in common browsers on macOS, Windows, or Linux and needs a stable internet connection. CATMA Standalone can run on a local machine or private network, including settings without internet access. Projects can be exported as text, TEI-XML, or CSV; data is also accessible through Git, the GitMA Python package, or the JSON API. The hosted service stores data in Germany and makes automatic backups at least daily. CATMA says its server data is not encrypted at rest, though off-site backups are encrypted, and advises against uploading sensitive data. Standalone defaults to single-user use; some features need additional configuration or are unavailable.