Word Sense Disambiguation is a longstanding task in Natural Language Processing, lying at the core of human language understanding. However, the evaluation of automatic systems has been problematic, mainly due to the lack of a reliable evaluation framework. In this paper we develop a unified evaluation framework and analyze the performance of various Word Sense Disambiguation systems in a fair setup. The results show that supervised systems clearly outperform knowledge-based models. Among the supervised systems, a linear classi- fier trained on conventional local features still proves to be a hard baseline to beat. Nonetheless, recent approaches exploiting neural networks on unlabeled corpora achieve promising results, surpassing this hard baseline in most test sets.

Raganato, A., Camacho-collados, J., Navigli, R. (2017). Word sense disambiguation: a uinified evaluation framework and empirical comparison. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers (pp.99-110). Association for Computational Linguistics [10.18653/v1/e17-1010].

Word sense disambiguation: a uinified evaluation framework and empirical comparison

Raganato, Alessandro
;
2017

Abstract

Word Sense Disambiguation is a longstanding task in Natural Language Processing, lying at the core of human language understanding. However, the evaluation of automatic systems has been problematic, mainly due to the lack of a reliable evaluation framework. In this paper we develop a unified evaluation framework and analyze the performance of various Word Sense Disambiguation systems in a fair setup. The results show that supervised systems clearly outperform knowledge-based models. Among the supervised systems, a linear classi- fier trained on conventional local features still proves to be a hard baseline to beat. Nonetheless, recent approaches exploiting neural networks on unlabeled corpora achieve promising results, surpassing this hard baseline in most test sets.
paper
Semantics; Knowledge based systems; disambiguation WSD
English
15th Conference of the European Chapter of the Association for Computational Linguistics
2017
Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers
978-1-945626-34-0
2017
1
99
110
http://aclweb.org/anthology/E/E17/E17-1010.pdf
reserved
Raganato, A., Camacho-collados, J., Navigli, R. (2017). Word sense disambiguation: a uinified evaluation framework and empirical comparison. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers (pp.99-110). Association for Computational Linguistics [10.18653/v1/e17-1010].
File in questo prodotto:
File Dimensione Formato  
Raganato_Word_2017.pdf

Solo gestori archivio

Dimensione 275.96 kB
Formato Adobe PDF
275.96 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/361554
Citazioni
  • Scopus 249
  • ???jsp.display-item.citation.isi??? ND
Social impact