Evaluating LSTM models for grammatical function labelling
To improve grammatical function labelling for German, we augment the labelling component of a neural dependency parser with a decision history. We present different ways to encode the history, using different LSTM architectures, and show that our models yield significant improvements, resulting in a...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Dokumenttyp: | Kapitel/Artikel Konferenzschrift |
| Sprache: | Englisch |
| Veröffentlicht: |
September 2017
|
| In: |
15th International Conference on Parsing Technologies - proceedings of the conference
Year: 2017, Pages: 128-133 |
| Online-Zugang: | Verlag, lizenzpflichtig, Volltext: https://aclanthology.org/W17-6318
|
| Verfasserangaben: | Bich-Ngoc Do, Ines Rehbein |
| Zusammenfassung: | To improve grammatical function labelling for German, we augment the labelling component of a neural dependency parser with a decision history. We present different ways to encode the history, using different LSTM architectures, and show that our models yield significant improvements, resulting in a LAS for German that is close to the best result from the SPMRL 2014 shared task (without the reranker). |
|---|---|
| Beschreibung: | Gesehen am 22.11.2023 |
| Beschreibung: | Online Resource |
| ISBN: | 9781945626739 |