0440949 Andreas van Cranenburgh
Cognitive Models of Language research proposal

Enriching DOP with morphology: a pilot using a constructed language

Esperanto is a constructed language with a rich and regular morphology,
hence it is likely that taking the latter into account will improve parse
accuracy. However, whether morphological and phrase structure analysis
can be considered as separate, autonomous steps, or should rather
be combined into a single DOP model is an empirical question which I would like
to answer.

Since there is no gold standard treebank with phrase structures for Esperanto,
I will construct a small toy corpus for testing. Furthermore, experiments with
U-Dop for both morphology and phrase structure are possible.

Lastly, previous work with Esperanto has resulted in two highly successful
(>95% precision on a small test corpus) dependency grammars (Bick 2007, Gobbo
2009), a constraint grammar and an adpositional grammar, respectively, which
provides a means of comparison and, if it is possible to make DOP
use dependency structures, a potential treebank.