Transformer based named entity recognition for place name extraction from unstructured text

Cillian Berragan; Alex Singleton; Alessia Calafiore; Jeremy Morley

doi:10.1080/13658816.2022.2133125

Transformer based named entity recognition for place name extraction from unstructured text

Cillian Berragan^*, Alex Singleton, Alessia Calafiore, Jeremy Morley

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

Abstract / Description of output

Place names embedded in online natural language text present a useful source of geographic information. Despite this, many methods for the extraction of place names from text use pre-trained models that were not explicitly designed for this task. Our paper builds five custom-built Named Entity Recognition (NER) models and evaluates them against three popular pre-built models for place name extraction. The models are evaluated using a set of manually annotated Wikipedia articles with reference to the F₁ score metric. Our best performing model achieves an F₁ score of 0.939 compared with 0.730 for the best performing pre-built model. Our model is then used to extract all place names from Wikipedia articles in Great Britain, demonstrating the ability to more accurately capture unknown place names from volunteered sources of online geographic information.

Original language	English
Pages (from-to)	1-20
Number of pages	20
Journal	International Journal of Geographical Information Science
Early online date	17 Oct 2022
DOIs	https://doi.org/10.1080/13658816.2022.2133125
Publication status	E-pub ahead of print - 17 Oct 2022

Keywords / Materials (for Non-textual outputs)

named entity recognition
natural language processing
place name extraction
volunteered geographic information

Access to Document

10.1080/13658816.2022.2133125Licence: Creative Commons: Attribution (CC-BY)

Calafiore2022IJGISTransformerFinal published version, 2.74 MBLicence: Creative Commons: Attribution (CC-BY)

Cite this

@article{674800f7c3f4411aa4b396991d0c2e4d,

title = "Transformer based named entity recognition for place name extraction from unstructured text",

abstract = "Place names embedded in online natural language text present a useful source of geographic information. Despite this, many methods for the extraction of place names from text use pre-trained models that were not explicitly designed for this task. Our paper builds five custom-built Named Entity Recognition (NER) models and evaluates them against three popular pre-built models for place name extraction. The models are evaluated using a set of manually annotated Wikipedia articles with reference to the F1 score metric. Our best performing model achieves an F1 score of 0.939 compared with 0.730 for the best performing pre-built model. Our model is then used to extract all place names from Wikipedia articles in Great Britain, demonstrating the ability to more accurately capture unknown place names from volunteered sources of online geographic information.",

keywords = "named entity recognition, natural language processing, place name extraction, volunteered geographic information",

author = "Cillian Berragan and Alex Singleton and Alessia Calafiore and Jeremy Morley",

year = "2022",

month = oct,

day = "17",

doi = "10.1080/13658816.2022.2133125",

language = "English",

pages = "1--20",

journal = "International Journal of Geographical Information Science",

issn = "1365-8816",

publisher = "Taylor & Francis",

}

TY - JOUR

T1 - Transformer based named entity recognition for place name extraction from unstructured text

AU - Berragan, Cillian

AU - Singleton, Alex

AU - Calafiore, Alessia

AU - Morley, Jeremy

PY - 2022/10/17

Y1 - 2022/10/17

N2 - Place names embedded in online natural language text present a useful source of geographic information. Despite this, many methods for the extraction of place names from text use pre-trained models that were not explicitly designed for this task. Our paper builds five custom-built Named Entity Recognition (NER) models and evaluates them against three popular pre-built models for place name extraction. The models are evaluated using a set of manually annotated Wikipedia articles with reference to the F1 score metric. Our best performing model achieves an F1 score of 0.939 compared with 0.730 for the best performing pre-built model. Our model is then used to extract all place names from Wikipedia articles in Great Britain, demonstrating the ability to more accurately capture unknown place names from volunteered sources of online geographic information.

AB - Place names embedded in online natural language text present a useful source of geographic information. Despite this, many methods for the extraction of place names from text use pre-trained models that were not explicitly designed for this task. Our paper builds five custom-built Named Entity Recognition (NER) models and evaluates them against three popular pre-built models for place name extraction. The models are evaluated using a set of manually annotated Wikipedia articles with reference to the F1 score metric. Our best performing model achieves an F1 score of 0.939 compared with 0.730 for the best performing pre-built model. Our model is then used to extract all place names from Wikipedia articles in Great Britain, demonstrating the ability to more accurately capture unknown place names from volunteered sources of online geographic information.

KW - named entity recognition

KW - natural language processing

KW - place name extraction

KW - volunteered geographic information

UR - http://www.scopus.com/inward/record.url?scp=85140127198&partnerID=8YFLogxK

U2 - 10.1080/13658816.2022.2133125

DO - 10.1080/13658816.2022.2133125

M3 - Article

AN - SCOPUS:85140127198

SN - 1365-8816

SP - 1

EP - 20

JO - International Journal of Geographical Information Science

JF - International Journal of Geographical Information Science

ER -

Transformer based named entity recognition for place name extraction from unstructured text

Abstract / Description of output

Keywords / Materials (for Non-textual outputs)

Access to Document

Other files and links

Fingerprint

Cite this