We present an evaluation of three anaphora resolution algorithms – GuiTAR, JavaRAP and MARS – based on a subpart of the Susanne Corpus. The results were then compared to the GETARUNS system developed at Department of Language Sciences, Università Ca’ Foscari, Venezia, by prof. Delmonte. Our aim was to highlight differences between the systems and to see if knowledge-poor approach- es are suitable for AR, as they are much less labour-intensive and time-consum- ing than knowledge-rich systems.
Knowledge-poor and Knowledge-rich Approach in Anaphora Resolution Algorithms : a Comparison
DELMONTE, Rodolfo;
2009-01-01
Abstract
We present an evaluation of three anaphora resolution algorithms – GuiTAR, JavaRAP and MARS – based on a subpart of the Susanne Corpus. The results were then compared to the GETARUNS system developed at Department of Language Sciences, Università Ca’ Foscari, Venezia, by prof. Delmonte. Our aim was to highlight differences between the systems and to see if knowledge-poor approach- es are suitable for AR, as they are much less labour-intensive and time-consum- ing than knowledge-rich systems.
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
2428617_SAM.pdf
accesso aperto
Tipologia:
Abstract
Licenza:
Licenza non definita
Dimensione
103.05 kB
Formato
Adobe PDF
|
103.05 kB | Adobe PDF | Visualizza/Apri |
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
