SATEK

This DFG-funded project (project number: 531750631) develops new methods for the thematic classification of very large text corpora in the digital humanities and social sciences. Focusing on the German Reference Corpus (DeReKo), the world’s largest collection of German-language texts, it addresses the lack of reliable topic-based metadata for heterogeneous and rapidly growing corpora. By combining approaches from computer science and corpus linguistics, the project creates efficient, open-source, and dynamic classification methods that support advanced corpus analysis, stratified sampling, and the study of linguistic variation. The methods are also tested on domain-specific resources, such as Grammis, to enable fine-grained thematic indexing of specialized texts.

Project Locations

The project is carried out at three research institutions:

Team Frankfurt

Publications

Total: 1
Bhuvanesh Verma and Alexander Mehler. 2026. Predicting Topic (Co-)Occurrence Using Topic Networks Built from the Project Gutenberg Corpus. Proceedings of the 15th International Conference on Language Resources and Evaluation (LREC 2026). accepted.
BibTeX 
@inproceedings{Verma:Mehler:2026,
  title     = {Predicting Topic (Co-)Occurrence Using Topic Networks Built from
               the Project Gutenberg Corpus},
  booktitle = {Proceedings of the 15th International Conference on Language Resources
               and Evaluation (LREC 2026)},
  year      = {2026},
  author    = {Verma, Bhuvanesh and Mehler, Alexander},
  keywords  = {Topic Evolution, Topic Network,Time-aware Networks, Temporal Autocorrelation, Project Gutenberg, satek},
  abstract  = {Although temporal topic modeling has been widely applied to scientific
               and legal texts, literary corpora have largely been overlooked
               in this regard. To address this issue, we analyze topic evolution
               in a subset of the Project Gutenberg (PG) corpus. We model this
               subset as a sequence of topic networks that capture the emergence,
               persistence, and interaction of thematic structures over decades.
               Using supervised topic representations, we predict nodes (topics)
               and edges (topic pairings) to forecast future topics and their
               co-occurrence. Our experiments demonstrate moderate to strong
               temporal persistence in topic connectivity patterns across three
               topic systems, with ROC-AUC and AP values consistently above 0.85.
               We find that the temporal span of topic networks significantly
               impacts predictive performance: longer spans improve the stability
               and recall of topic presence, while shorter spans better capture
               evolving topic relationships. Overall, our findings demonstrate
               the predictability of topics in literary texts over time.},
  note      = {accepted}
}