GramLab: Multilingual Treebank-Based Deep LFG Grammars for Parsing, Generation and Machine Translation
Welcome to the website of the GramLab Research Group in the National Centre for Language Technology (NCLT) and the Centre for Next Generation Localisation (CNGL).
We use treebanks to automatically acquire deep Lexical-Functional Grammar (LFG)
resources for wide-coverage, robust, statistical, multi-lingual parsing and generation.
To date, we have required resources for English, Chinese, Arabic, Japanese, German,
French, and Spanish. Our research includes:
- Treebank-Based LFG Grammar and Lexicon Acquisition
- Generation/Surface Realisation
- Machine Translation Evaluation
|We also use hand-crafted ParGram LFG grammars in some of our research on Machine Translation and Grammatical Error Detection.|
|Thanks for accessing the GramLab site!
|-- Josef van Genabith, October 2008||
| Recent News |
|October 2010: Yvette Graham successfully defended her PhD thesis entitled "Deep Syntax in Statistical Machine Translation". Congratulations Yvette!|
|July 2010: Congratulations to Natalie Schluter who successfully defended her PhD thesis entitled "Treebank-Based Deep Grammar Acquisition for French Probabilistic Parsing Resources".|
|July 2009: Congratulations to Ines Rehbein who successfully defended her PhD thesis entitled "Treebank-Based Grammar
Acquisition for German" and to Yuqing Guo who successfully defended her PhD thesis entitled "Treebank-based Acquisition of Chinese LFG Resources for Parsing and Generation".|