Comparing Annotated Datasets for Named Entity Recognition in English Literature

Rositsa V. Ivanova; Sabrina Kirrane; Marieke van Erp

Comparing Annotated Datasets for Named Entity Recognition in English Literature

Rositsa V. Ivanova, Sabrina Kirrane, Marieke van Erp

Onderzoeksoutput: Bijdrage aan conferentie › Paper › Wetenschappelijk › peer review

3 Citaten (Scopus)

Samenvatting

The growing interest in named entity recognition (NER) in various domains has led to the creation of different benchmark datasets, often with slightly different annotation guidelines. To better understand the different NER benchmark datasets for the domain of English literature and their impact on the evaluation of NER tools, we analyse two existing annotated datasets and create two additional gold standard datasets. Following on from this, we evaluate the performance of two NER tools, one domain-specific and one general-purpose NER tool, using the four gold standards, and analyse the sources for the differences in the measured performance. Our results show that the performance of the two tools varies significantly depending on the gold standard used for the individual evaluations.

Originele taal-2	Engels
Pagina's	3788-3797
Aantal pagina's	10
Status	Gepubliceerd - 2022

Toegang tot document

http://www.lrec-conf.org/proceedings/lrec2022/pdf/2022.lrec-1.404.pdf

Andere bestanden en links

Link naar publicatie in Scopus

Citeer dit

@conference{57336e2d8afe4c6094313236d20ec424,

title = "Comparing Annotated Datasets for Named Entity Recognition in English Literature",

abstract = "The growing interest in named entity recognition (NER) in various domains has led to the creation of different benchmark datasets, often with slightly different annotation guidelines. To better understand the different NER benchmark datasets for the domain of English literature and their impact on the evaluation of NER tools, we analyse two existing annotated datasets and create two additional gold standard datasets. Following on from this, we evaluate the performance of two NER tools, one domain-specific and one general-purpose NER tool, using the four gold standards, and analyse the sources for the differences in the measured performance. Our results show that the performance of the two tools varies significantly depending on the gold standard used for the individual evaluations.",

keywords = "annotated datasets, annotation guidelines, corpus annotation, literature, named entity recognition",

author = "Ivanova, {Rositsa V.} and Sabrina Kirrane and {van Erp}, Marieke",

note = "Funding Information: We thank Martin Beno and Martin Trajkov for the annotation of the texts. Publisher Copyright: {\textcopyright} European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0.",

year = "2022",

language = "English",

pages = "3788--3797",

}

TY - CONF

T1 - Comparing Annotated Datasets for Named Entity Recognition in English Literature

AU - Ivanova, Rositsa V.

AU - Kirrane, Sabrina

AU - van Erp, Marieke

N1 - Funding Information: We thank Martin Beno and Martin Trajkov for the annotation of the texts. Publisher Copyright: © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0.

PY - 2022

Y1 - 2022

N2 - The growing interest in named entity recognition (NER) in various domains has led to the creation of different benchmark datasets, often with slightly different annotation guidelines. To better understand the different NER benchmark datasets for the domain of English literature and their impact on the evaluation of NER tools, we analyse two existing annotated datasets and create two additional gold standard datasets. Following on from this, we evaluate the performance of two NER tools, one domain-specific and one general-purpose NER tool, using the four gold standards, and analyse the sources for the differences in the measured performance. Our results show that the performance of the two tools varies significantly depending on the gold standard used for the individual evaluations.

AB - The growing interest in named entity recognition (NER) in various domains has led to the creation of different benchmark datasets, often with slightly different annotation guidelines. To better understand the different NER benchmark datasets for the domain of English literature and their impact on the evaluation of NER tools, we analyse two existing annotated datasets and create two additional gold standard datasets. Following on from this, we evaluate the performance of two NER tools, one domain-specific and one general-purpose NER tool, using the four gold standards, and analyse the sources for the differences in the measured performance. Our results show that the performance of the two tools varies significantly depending on the gold standard used for the individual evaluations.

KW - annotated datasets

KW - annotation guidelines

KW - corpus annotation

KW - literature

KW - named entity recognition

UR - http://www.scopus.com/inward/record.url?scp=85144442352&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85144442352

SP - 3788

EP - 3797

ER -