Publication at Faculty of Mathematics and Physics |
2016
Abstract
We present a design for a new electronic lexicon of Czech discourse connectives. The data format and the annotation scheme are based on a study of similar existing resources, and we discuss arguments for choosing the data structure and selecting features of the lexicon entries.
A special attention is paid to a consistent encoding of both primary and secondary connectives. The data itself comes from exploiting the Prague Dependency Treebank, a large treebank manually annotated with discourse relations.