Publication at Faculty of Mathematics and Physics |
2022
Abstract
A theoretically well-founded annotation of language corpora is a very importatnt component part of such corpora. In our paper, we support this point by the example of the Prague Dependency Treebank and its annotation scheme.
We bring some case studies concerning the information structure of the sentence. At the same time, we also demonstrate that the work with a consistently annotated corpus may lead to some modifications of the original theory.