Syntax-based Discourse Segmentation of Dutch Text
N.H. van der Vliet, Marija Slavkovik · University of Groningen research database (University of Groningen / Centre for Information Technology) · 2010
An essential step in discourse parsing is the identification of suitable Elementary Discourse Units (EDUs). For English and German written text several automatic discourse segmentation approaches have been developed. For Dutch, far less research is available. This paper presents an approach to automatically decompose Dutch written text into EDUs, based on syntactic information. We describe discourse segmentation guidelines for Dutch, the discourse segmentation algorithm, and the results of applying this algorithm to an annotated corpus.