Detecting Anaphoricity and Antecedenthood for Coreference Resolution
Most coreference resolution (CR) systems try to resolve all “candidate anaphors” by comparing them to all preceding “candidate antecedents” until the correct one is found. In this study we investigate possibilities to identify unlikely anaphors and antecedents. We evaluate our approach on the MUC-7...
Bewaard in:
| Gepubliceerd in: | Procesamiento del Lenguaje Natural |
|---|---|
| Hoofdauteur: | |
| Formaat: | Artigo |
| Taal: | Inglês |
| Gepubliceerd in: |
Sociedad Española para el Procesamiento del Lenguaje Natural
2009
|
| Onderwerpen: | |
| Online toegang: | https://www.redalyc.org/articulo.oa?id=515751742014 |
| Tags: |
Geen labels, Wees de eerste die dit record labelt!
|