Arc-Eager Parsing with the Tree Constraint
Bibliographic Data
| ID | 12155648 |
|---|---|
| Authors | Joakim Nivre (0000-0002-7873-3971, Uppsala University, corresponding author), Daniel Fernández‐González (0000-0002-6733-2371, Universidade de Vigo), Daniel Fernández-González (Universidade de Vigo) |
| Year | 2014 |
| Volume | 40 |
| Issue | 2 |
| Pages | 259-267 |
| Publication date | 2014-02-24 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Computational Linguistics (JOURNAL) |
| Journal identifiers | ISSN: 0891-2017 • E-ISSN: 1530-9312 |
| Publisher | Association for Computational Linguistics (PUBLISHER • US) |
| DOI | 10.1162/coli_a_00185 |
| OpenAlex | W2121293489 |
| Language | EN |
| References cited | 7 |
The arc-eager system for transition-based dependency parsing is widely used in natural language processing despite the fact that it does not guarantee that the output is a well-formed dependency tree. We propose a simple modification to the original system that enforces the tree constraint without requiring any modification to the parser training procedure. Experiments on multiple languages show that the method on average achieves 72% of the error reduction possible and consistently outperforms the standard heuristic in current use
Algorithm · Constraint (computer-aided design · Dependency (UML · Dependency grammar · Heuristic · Natural language processing · Parsing · Reduction (mathematics · Tree (set theory · Computer Science · Mathematics · Natural Language Processing Techniques · Semantic Web and Ontologies · Topic Modeling · Artificial Intelligence · Theoretical Computer Science
| Highly cited | No |
|---|