KorAP is a corpus analysis platform for very large, multiply annotated corpora and complex search queries. It is developed at the Leibniz Institute for the German Language (IDS) and serves the German Reference Corpus DeReKo at korap.ids-mannheim.de.
With Docker (or rootless Podman) installed, this command starts KorAP with an example corpus:
curl -s https://raw.githubusercontent.com/KorAP/KorAP-Docker/master/compose.yaml | \
COMPOSE_PROFILES=open,example INDEX=example-index docker compose -p korap -f - upThen open http://localhost:64543 and search for geht.
KorAP-Ingestion takes your TEI XML files, annotates and indexes them,
and starts a KorAP instance on the result – by just typing make.
To run KorAP on an existing index or with login, see KorAP-Docker.
- RKorAPClient – R package for queries, frequencies and collocations
- PythonKorAPClient – the same for Python
| Repository | Purpose | |
|---|---|---|
| 🐙 | Kalamar | Web user interface |
| 🚤 | Kustvakt | API, user and access management |
| 🔍 | Krill | Lucene-based search index |
| 📝 | Koral | Query language translation to KoralQuery |
| 🐳 | KorAP-Docker | Run all components with one command |
Conversion and annotation tools: KorAP-XML-TEI · korapxmltool · KorAP-Tokenizer · Datok
Found a problem or have a question? Open an issue at Kalamar – we will take it from there.