Multilingual semantic role labeling
Björkelund, Anders LU ; Hafdell, Love and Nugues, Pierre LU
(2009)
p.43-48
- Abstract
- This paper describes our contribution to the semantic role labeling task (SRL-only) of the CoNLL-2009 shared task in the closed challenge(Hajic et al., 2009). Our system consists of a pipeline of independent, local classifiers
that identify the predicate sense, the arguments of the predicates, and the argument labels. Using these local models, we carried out a beam search to generate a pool of candidates. We then reranked the candidates using a joint learning approach that combines the local models and proposition features.
To address the multilingual nature of the data, we implemented a feature selection procedure that systematically explored the feature space, yielding significant gains over a standard set of features.... (More) - This paper describes our contribution to the semantic role labeling task (SRL-only) of the CoNLL-2009 shared task in the closed challenge(Hajic et al., 2009). Our system consists of a pipeline of independent, local classifiers
that identify the predicate sense, the arguments of the predicates, and the argument labels. Using these local models, we carried out a beam search to generate a pool of candidates. We then reranked the candidates using a joint learning approach that combines the local models and proposition features.
To address the multilingual nature of the data, we implemented a feature selection procedure that systematically explored the feature space, yielding significant gains over a standard set of features. Our system achieved the second best semantic score overall with an average labeled semantic F1 of 80.31. It obtained the best F1 score on the Chinese and German data
and the second best one on English. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/1668770
- author
- Björkelund, Anders
LU
; Hafdell, Love
and Nugues, Pierre
LU
- organization
- publishing date
- 2009
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Proceedings of The Thirteenth Conference on Computational Natural Language Learning (CoNLL-2009)
- pages
- 43 - 48
- external identifiers
-
- scopus:84859994270
- language
- English
- LU publication?
- yes
- id
- 7e94d684-fdf1-4e6c-9522-ac8d9ea1f7e1 (old id 1668770)
- alternative location
- http://www.aclweb.org/anthology/W/W09/W09-1206.pdf
- date added to LUP
- 2016-04-04 14:05:03
- date last changed
- 2022-01-30 01:23:30
@inproceedings{7e94d684-fdf1-4e6c-9522-ac8d9ea1f7e1,
abstract = {{This paper describes our contribution to the semantic role labeling task (SRL-only) of the CoNLL-2009 shared task in the closed challenge(Hajic et al., 2009). Our system consists of a pipeline of independent, local classifiers<br/><br>
that identify the predicate sense, the arguments of the predicates, and the argument labels. Using these local models, we carried out a beam search to generate a pool of candidates. We then reranked the candidates using a joint learning approach that combines the local models and proposition features.<br/><br>
To address the multilingual nature of the data, we implemented a feature selection procedure that systematically explored the feature space, yielding significant gains over a standard set of features. Our system achieved the second best semantic score overall with an average labeled semantic F1 of 80.31. It obtained the best F1 score on the Chinese and German data<br/><br>
and the second best one on English.}},
author = {{Björkelund, Anders and Hafdell, Love and Nugues, Pierre}},
booktitle = {{Proceedings of The Thirteenth Conference on Computational Natural Language Learning (CoNLL-2009)}},
language = {{eng}},
pages = {{43--48}},
title = {{Multilingual semantic role labeling}},
url = {{http://www.aclweb.org/anthology/W/W09/W09-1206.pdf}},
year = {{2009}},
}