Extended constituent-to-dependency conversion for English
Författare
Redaktör
- Joakim Nivre
- Heiki-Jaan Kalep
- Kadri Muischnek
- Mare Koit
Summary, in English
The resulting trees generally have a more complex dependency structure. For example, 6% of the trees contain at least one nonprojective link, which is difficult for many parsing algorithms. As can be expected, the more complex structure and the enriched set of edge labels make the trees more difficult to predict, and we observed a decrease in parsing accuracy when applying two dependency parsers to the new corpus. However, the richer information contained in the new trees resulted in a 23% error reduction in a baseline FrameNet semantic role labeler that relied on dependency arc labels only.
Publiceringsår
2007
Språk
Engelska
Sidor
105-112
Publikation/Tidskrift/Serie
NODALIDA 2007 Proceedings
Fulltext
- Available as PDF - 511 kB
- Download statistics
Länkar
Dokumenttyp
Konferensbidrag
Förlag
University of Tartu
Ämne
- Computer Science
Nyckelord
- dependency syntax
- treebanks
- Natural language processing
Conference name
16th Nordic Conference of Computational Linguistics
Conference date
2007-05-25 - 2007-05-26
Conference place
Tartu, Estonia
Status
Published
ISBN/ISSN/Övrigt
- ISBN: 978-9985-4-0514-7