Extending a multilingual lexical resource by bootstrapping named entity classification using Wikipedia's category system

Knopp, Johannes

Vorschau

PDF
Knopp11ExtendingHeiNER.pdf - Veröffentlichte Version
Download (345kB)

URL:	https://madoc.bib.uni-mannheim.de/29542
Weitere URL:	https://www.aclweb.org/anthology/W11-3607/
URN:	urn:nbn:de:bsz:180-madoc-295426
Dokumenttyp:	Konferenzveröffentlichung
Erscheinungsjahr:	2011
Buchtitel:	Proceedings of the Fifth International Workshop On Cross Lingual Information Access
Seitenbereich:	35-43
Veranstaltungstitel:	Fifth International Workshop On Cross Lingual Information Access
Veranstaltungsort:	Chiang Mai, Thailand
Veranstaltungsdatum:	8.-13. Nov 2011
Ort der Veröffentlichung:	Chiang Mai, Thailand
Verlag:	Asian Federation of Natural Language Processing
Sprache der Veröffentlichung:	Englisch
Einrichtung:	Fakultät für Wirtschaftsinformatik und Wirtschaftsmathematik > Practical Computer Science II: Artificial Intelligence (Stuckenschmidt 2009-)
Fachgebiet:	004 Informatik
Fachklassifikation:	CCS:
Freie Schlagwörter (Deutsch):	Named Entities, Wikipedia, HeiNER, NERC
Abstract:	Named Entity Recognition and Classification (NERC) is a well-studied NLP task which is typically approached using machine learning algorithms that rely on training data whose creation usually is expensive. The high costs result in the lack of NERC training data for many languages. An approach to create a multilingual NE corpus was presented in Wentland et al. (2008). The resulting resource called HeiNER describes a valuable number of NEs but does not include their types. We present a bootstrap approach based on Wikipedia’s category system to classify the NEs contained in HeiNER that is able to classify more than two million named entities to improve the resource’s quality.
Zusätzliche Informationen:	Online-Ressource