Head selection parsers and LSTM labelers
This resource contains code, data and pre-trained models for various types of neural dependency parsers and LSTM labelers used in the papers: Do et al. (2017). "What Do We Need to Know About an Unknown Word When Parsing German" Do and Rehbein (2017). "Evaluating LSTM Models for Gramma...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Database Research Data |
| Language: | English |
| Published: |
Heidelberg
Universität
2023-11-13
|
| DOI: | 10.11588/data/BPWWJL |
| Subjects: | |
| Online Access: | Resolving-System, kostenfrei, Volltext: https://doi.org/10.11588/data/BPWWJL Verlag, kostenfrei, Volltext: https://heidata.uni-heidelberg.de/dataset.xhtml?persistentId=doi:10.11588/data/BPWWJL 
|
| Author Notes: | Bich-Ngoc Do, Ines Rehbein, Anette Frank |
| Summary: | This resource contains code, data and pre-trained models for various types of neural dependency parsers and LSTM labelers used in the papers: Do et al. (2017). "What Do We Need to Know About an Unknown Word When Parsing German" Do and Rehbein (2017). "Evaluating LSTM Models for Grammatical Function Labelling" The parsers and labelers are inspired by the head-selection parser of Zhang et al., (2017). We extend the parser to use different input features, namely: Word embeddings POS tag embeddings Constituent embeddings (e.g., characters or compound) and their combinations. Grammatical function labeling is formulated as a sequence labeling task. We introduce two new bidirectional LSTMs labelers with different orders of tree nodes (linear and BFS order) and another labeler based on top-down tree LSTMs. |
|---|---|
| Item Description: | Produktionsdatum: 2017 Gesehen am 22.11.2023 |
| Physical Description: | Online Resource |
| DOI: | 10.11588/data/BPWWJL |