Abstract
In this paper we demonstrate a hybrid treebank encoding format, derived from the dependency-based format used in Prague Dependency Treebank (PDT). We have specified a Prague Markup Language (PML) profile for the SemTi- Kamols hybrid grammar model that has been developed for languages with relatively free word order (e.g. Latvian). This has allowed us to exploit the tree editor TrEd that has been used in PDT development. As a proof of concept, a small Latvian treebank has been created by annotating 100 sentences from "Sophie's World".
| Original language | English |
|---|---|
| Pages | 303-306 |
| Number of pages | 4 |
| Publication status | Published - 2011 |
| Externally published | Yes |
| Event | 18th Nordic Conference of Computational Linguistics, NODALIDA 2011 - Riga, Latvia Duration: 11 May 2011 → 13 May 2011 |
Conference
| Conference | 18th Nordic Conference of Computational Linguistics, NODALIDA 2011 |
|---|---|
| Country/Territory | Latvia |
| City | Riga |
| Period | 11/05/11 → 13/05/11 |
Fingerprint
Dive into the research topics of 'A prague markup language profile for the SemTi-Kamols grammar model'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver