Proceedings of the Workshop on Coreference Resolution Beyond OntoNotes (CORBON 2016) 2016
DOI: 10.18653/v1/w16-0710
|View full text |Cite
|
Sign up to set email alerts
|

Coreference Resolution for the Basque Language with BART

Abstract: In this paper we present our work on Coreference Resolution in Basque, a unique language which poses interesting challenges for the problem of coreference. We explain how we extend the coreference resolution toolkit, BART, in order to enable it to process Basque. Then we run four different experiments showing both a significant improvement by extending a baseline feature set and the effect of calculating performance of hand-parsed mentions vs. automatically parsed mentions. Finally, we discuss some key charact… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
2
1

Year Published

2019
2019
2019
2019

Publication Types

Select...
2
1

Relationship

2
1

Authors

Journals

citations
Cited by 3 publications
(3 citation statements)
references
References 10 publications
0
2
1
Order By: Relevance
“…Moreover, to contextualize the results we obtained, in Table 3, we can see the results of the neural cross-lingual system in comparison with previous coreference resolution systems for Basque. The results obtained are lower than those obtained by previous rule-based (Soraluze et al, 2015) and ML-based (Soraluze et al, 2016) In Table 5 we can see an example of the type of mistakes in the output of our cross-lingual system. Key refers to gold annotation and response to the output of the system.…”
Section: Resultscontrasting
confidence: 62%
See 1 more Smart Citation
“…Moreover, to contextualize the results we obtained, in Table 3, we can see the results of the neural cross-lingual system in comparison with previous coreference resolution systems for Basque. The results obtained are lower than those obtained by previous rule-based (Soraluze et al, 2015) and ML-based (Soraluze et al, 2016) In Table 5 we can see an example of the type of mistakes in the output of our cross-lingual system. Key refers to gold annotation and response to the output of the system.…”
Section: Resultscontrasting
confidence: 62%
“…As regards the Basque language, this is the first work about neural coreference resolution. Nevertheless, a rule-based coreference resolution system (Soraluze et al, 2015) and a machine learning based system (Soraluze et al, 2016) have been developed. Both of which used a rule-based mention detector (Soraluze et al, 2017).…”
Section: Related Workmentioning
confidence: 99%
“…There is a large existing body of work on coreference resolution spanning from the 1970s of which Poesio et al (2016) provides an exhaustive review. In recent years, the field has been dominated by machine learning approaches -with the notable exception of the rule-based system of H. Wiseman et al 2016), gradually reducing their dependency on rich features coming from preprocessing pipelines using linguistic knowledge.…”
Section: Related Workmentioning
confidence: 99%