Robust multilingual Named Entity Recognition with shallow semi-supervised features

We present a multilingual Named Entity Recognition approach based on a robust and general set of features across languages and datasets. Our system combines shallow local information with clustering semi-supervised features induced on large amounts of unlabeled text. Understanding via empirical expe...

Descripción completa

Detalles Bibliográficos
Autores: Agerri, Rodrigo, Rigau Claramunt, German
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2016
País:España
Institución:Universitat Pompeu Fabra
Repositorio:Repositorio Digital de la UPF
OAI Identifier:oai:repositori.upf.edu:10230/33529
Acceso en línea:http://hdl.handle.net/10230/33529
http://dx.doi.org/10.1016/j.artint.2016.05.003
Access Level:acceso abierto
Palabra clave:Named entity recognition
Information extraction
Clustering
Semi-supervised learning
Natural language processing
id ES_fc0c8656228bce220d337343bee7541a
oai_identifier_str oai:repositori.upf.edu:10230/33529
network_acronym_str ES
network_name_str España
repository_id_str
spelling Robust multilingual Named Entity Recognition with shallow semi-supervised featuresAgerri, RodrigoRigau Claramunt, GermanNamed entity recognitionInformation extractionClusteringSemi-supervised learningNatural language processingWe present a multilingual Named Entity Recognition approach based on a robust and general set of features across languages and datasets. Our system combines shallow local information with clustering semi-supervised features induced on large amounts of unlabeled text. Understanding via empirical experimentation how to effectively combine various types of clustering features allows us to seamlessly export our system to other datasets and languages. The result is a simple but highly competitive system which obtains state of the art results across five languages and twelve datasets. The results are reported on standard shared task evaluation data such as CoNLL for English, Spanish and Dutch. Furthermore, and despite the lack of linguistically motivated features, we also report best results for languages such as Basque and German. In addition, we demonstrate that our method also obtains very competitive results even when the amount of supervised data is cut by half, alleviating the dependency on manually annotated data. Finally, the results show that our emphasis on clustering features is crucial to develop robust out-of-domain models. The system and models are freely available to facilitate its use and guarantee the reproducibility of results.This work has been supported by the European projects NewsReader, EC/FP7/316404 and QTLeap – EC/FP7/610516, and by the Spanish Ministry of Economy and Competitiveness (MINECO) SKATER, Grant No. TIN2012-38584-C06-01 and TUNER, TIN2015-65308-C5-1-R.Elsevier20172016info:eu-repo/semantics/articleinfo:eu-repo/semantics/acceptedVersionapplication/pdfapplication/pdfhttp://hdl.handle.net/10230/33529http://dx.doi.org/10.1016/j.artint.2016.05.003reponame:Repositorio Digital de la UPFinstname:Universitat Pompeu FabraInglésArtificial Intelligence. 2016;238: 63-82.info:eu-repo/grantAgreement/EC/FP7/316404info:eu-repo/grantAgreement/EC/FP7/610516info:eu-repo/grantAgreement/ES/3PN/TIN2012-38584-C06-01info:eu-repo/grantAgreement/ES/1PE/TIN2015-65308-C5-1-R© Elsevier http://dx.doi.org/10.1016/j.artint.2016.05.003info:eu-repo/semantics/openAccessoai:repositori.upf.edu:10230/335292026-06-12T07:21:37Z
dc.title.none.fl_str_mv Robust multilingual Named Entity Recognition with shallow semi-supervised features
title Robust multilingual Named Entity Recognition with shallow semi-supervised features
spellingShingle Robust multilingual Named Entity Recognition with shallow semi-supervised features
Agerri, Rodrigo
Named entity recognition
Information extraction
Clustering
Semi-supervised learning
Natural language processing
title_short Robust multilingual Named Entity Recognition with shallow semi-supervised features
title_full Robust multilingual Named Entity Recognition with shallow semi-supervised features
title_fullStr Robust multilingual Named Entity Recognition with shallow semi-supervised features
title_full_unstemmed Robust multilingual Named Entity Recognition with shallow semi-supervised features
title_sort Robust multilingual Named Entity Recognition with shallow semi-supervised features
dc.creator.none.fl_str_mv Agerri, Rodrigo
Rigau Claramunt, German
author Agerri, Rodrigo
author_facet Agerri, Rodrigo
Rigau Claramunt, German
author_role author
author2 Rigau Claramunt, German
author2_role author
dc.subject.none.fl_str_mv Named entity recognition
Information extraction
Clustering
Semi-supervised learning
Natural language processing
topic Named entity recognition
Information extraction
Clustering
Semi-supervised learning
Natural language processing
description We present a multilingual Named Entity Recognition approach based on a robust and general set of features across languages and datasets. Our system combines shallow local information with clustering semi-supervised features induced on large amounts of unlabeled text. Understanding via empirical experimentation how to effectively combine various types of clustering features allows us to seamlessly export our system to other datasets and languages. The result is a simple but highly competitive system which obtains state of the art results across five languages and twelve datasets. The results are reported on standard shared task evaluation data such as CoNLL for English, Spanish and Dutch. Furthermore, and despite the lack of linguistically motivated features, we also report best results for languages such as Basque and German. In addition, we demonstrate that our method also obtains very competitive results even when the amount of supervised data is cut by half, alleviating the dependency on manually annotated data. Finally, the results show that our emphasis on clustering features is crucial to develop robust out-of-domain models. The system and models are freely available to facilitate its use and guarantee the reproducibility of results.
publishDate 2016
dc.date.none.fl_str_mv 2016
2017
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/acceptedVersion
format article
status_str acceptedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10230/33529
http://dx.doi.org/10.1016/j.artint.2016.05.003
url http://hdl.handle.net/10230/33529
http://dx.doi.org/10.1016/j.artint.2016.05.003
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Artificial Intelligence. 2016;238: 63-82.
info:eu-repo/grantAgreement/EC/FP7/316404
info:eu-repo/grantAgreement/EC/FP7/610516
info:eu-repo/grantAgreement/ES/3PN/TIN2012-38584-C06-01
info:eu-repo/grantAgreement/ES/1PE/TIN2015-65308-C5-1-R
dc.rights.none.fl_str_mv © Elsevier http://dx.doi.org/10.1016/j.artint.2016.05.003
info:eu-repo/semantics/openAccess
rights_invalid_str_mv © Elsevier http://dx.doi.org/10.1016/j.artint.2016.05.003
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:Repositorio Digital de la UPF
instname:Universitat Pompeu Fabra
instname_str Universitat Pompeu Fabra
reponame_str Repositorio Digital de la UPF
collection Repositorio Digital de la UPF
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869425385846865920
score 15,812429

Similares en LA Referencia