YAC – A Recursive Chunker for Unrestricted German Text
Hannah Kermes (Institute for Natural Language Processing, University of Stuttgart Azenbergstr. 12, 70174 Stuttgart, Germany)
Stefan Evert (Institute for Natural Language Processing, University of Stuttgart Azenbergstr. 12, 70174 Stuttgart, Germany)
WP5: Components & Systems
YAC is a fully automatic recursive chunker for unrestricted German text. It is especially designed to provide a useful basis for the extraction of linguistic as well as lexicographic information. Consequently, the grammar rules of YAC are implemented such as to make the resulting analysis meet the needs of an ensuing extraction process. The chunks provided by YAC are continuous parts of intra-clausal constituents including recursion but no PP-attachment or sentential elements. The chunks are additionally enriched with information about head lemma, morpho-syntactic features and certain lexical and structural properties.
Unrestricted german text