Ressourcen

Der SFB 1287 leistet einen Beitrag zur breiteren Forschungsgemeinschaft. Viele unserer Projekte haben Open-Access-Ressourcen wie Datenbanken und Modelle hervorgebracht, die von zukünftigen Forschern genutzt werden können und für die wissenschaftliche Gemeinschaft über den Rahmen des SFB 1287 hinaus von großer Bedeutung sind.

• A simplified version of the SWIFT model developed in Project B03 that can be run on a notebook computer in minutes. The relevant tools can be found in this tutorial paper.
• A questionnaire developed by Project C02 with structural and novel semantic diagnostics to probe the structure of serial verb constructions (Struck 2025).
• The Potsdamer Zeitungskorpus (Potsdam News Corpus) created by Project C06. This is a dependency treebank that includes newspapers from Early Modern German to Modern German and allows for the automatic measuring of linear distances between elements (Demske et al. 2025). It is accessible on TüNDRA.
• The online database CHAOS, created by Project C08 with their collected data on word order correlations in 40 typologically diverse languages.
• The corpus management tool MOCA3 (Multimodal Oral Corpora Administration 3), developed by Project C10. This is an online system for the administration of spoken language corpora that enables advanced searches and annotations.
• Paradise, the first aligned spoken-written dataset for German (including discourse annotations), created by Project T01 (Seemann et al. 2024a,b; Shahmohammadi & Stede 2024).
• RSTmulti is the largest corpus of multiple rhetorical text structure annotations. The corpus contains texts which have two separate Rhetorical Structure Theory annotations, for the purpose of studying disagreement in the context of discourse structure (Hewett & Stede 2025).