Abstract
Syntactic annotation of corpora is a useful corpus exploitation tool, presently limited to small corpora. The purpose of our project, entitled Syntactic Annotation of Czech Corpora, is to provide a large syntactically annotated corpus with customizable representation.
In this paper, I present the methods used for automatic syntactic annotation: a stochastic parser followed by a rule-based automatic correction module. The usefulness of such an annotation is demonstrated on frequency tables of syntactic functions of Czech nouns with respect to their case and preposition, which were extracted from the SYN2005 corpus, with additional syntactic annotation.