Training and Evaluation of Named Entity Recognition Models for Classical Latin. Beersmans, M., de Graaf, E., Van de Cruys, T., & Fantoli, M. In Proceedings of the Ancient Language Processing Workshop, pages 1–12, Varna, Bulgaria, 2023. INCOMA Ltd..
![Training and Evaluation of Named Entity Recognition Models for Classical Latin [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper abstract bibtex
Paper abstract bibtex We evaluate the performance of various models on the task of named entity recognition (NER) for classical Latin. Using an existing dataset, we train two transformer-based LatinBERT models and one shallow conditional random field (CRF) model. The performance is assessed using both standard metrics and a detailed manual error analysis, and compared to the results obtained by different already released Latin NER tools. Both analyses demonstrate that the BERT models achieve a better f1-score than the other models. Furthermore, we annotate new, unseen data for further evaluation of the models, and we discuss the impact of annotation choices on the results.
@inproceedings{beersmans_training_2023,
address = {Varna, Bulgaria},
title = {Training and {Evaluation} of {Named} {Entity} {Recognition} {Models} for {Classical} {Latin}},
url = {https://aclanthology.org/2023.alp-1.1.pdf},
abstract = {We evaluate the performance of various models on the task of named entity recognition (NER) for classical Latin. Using an existing dataset, we train two transformer-based LatinBERT models and one shallow conditional random field (CRF) model. The performance is assessed using both standard metrics and a detailed manual error analysis, and compared to the results obtained by different already released Latin NER tools. Both analyses demonstrate that the BERT models achieve a better f1-score than the other models. Furthermore, we annotate new, unseen data for further evaluation of the models, and we discuss the impact of annotation choices on the results.},
language = {English},
booktitle = {Proceedings of the {Ancient} {Language} {Processing} {Workshop}},
publisher = {INCOMA Ltd.},
author = {Beersmans, Marijke and de Graaf, Evelien and Van de Cruys, Tim and Fantoli, Margherita},
year = {2023},
pages = {1--12},
}Downloads: 0
{"_id":"eETwB7ridA8yAeyCE","bibbaseid":"beersmans-degraaf-vandecruys-fantoli-trainingandevaluationofnamedentityrecognitionmodelsforclassicallatin-2023","author_short":["Beersmans, M.","de Graaf, E.","Van de Cruys, T.","Fantoli, M."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"Varna, Bulgaria","title":"Training and Evaluation of Named Entity Recognition Models for Classical Latin","url":"https://aclanthology.org/2023.alp-1.1.pdf","abstract":"We evaluate the performance of various models on the task of named entity recognition (NER) for classical Latin. Using an existing dataset, we train two transformer-based LatinBERT models and one shallow conditional random field (CRF) model. The performance is assessed using both standard metrics and a detailed manual error analysis, and compared to the results obtained by different already released Latin NER tools. Both analyses demonstrate that the BERT models achieve a better f1-score than the other models. Furthermore, we annotate new, unseen data for further evaluation of the models, and we discuss the impact of annotation choices on the results.","language":"English","booktitle":"Proceedings of the Ancient Language Processing Workshop","publisher":"INCOMA Ltd.","author":[{"propositions":[],"lastnames":["Beersmans"],"firstnames":["Marijke"],"suffixes":[]},{"propositions":["de"],"lastnames":["Graaf"],"firstnames":["Evelien"],"suffixes":[]},{"propositions":["Van","de"],"lastnames":["Cruys"],"firstnames":["Tim"],"suffixes":[]},{"propositions":[],"lastnames":["Fantoli"],"firstnames":["Margherita"],"suffixes":[]}],"year":"2023","pages":"1–12","bibtex":"@inproceedings{beersmans_training_2023,\n\taddress = {Varna, Bulgaria},\n\ttitle = {Training and {Evaluation} of {Named} {Entity} {Recognition} {Models} for {Classical} {Latin}},\n\turl = {https://aclanthology.org/2023.alp-1.1.pdf},\n\tabstract = {We evaluate the performance of various models on the task of named entity recognition (NER) for classical Latin. Using an existing dataset, we train two transformer-based LatinBERT models and one shallow conditional random field (CRF) model. The performance is assessed using both standard metrics and a detailed manual error analysis, and compared to the results obtained by different already released Latin NER tools. Both analyses demonstrate that the BERT models achieve a better f1-score than the other models. Furthermore, we annotate new, unseen data for further evaluation of the models, and we discuss the impact of annotation choices on the results.},\n\tlanguage = {English},\n\tbooktitle = {Proceedings of the {Ancient} {Language} {Processing} {Workshop}},\n\tpublisher = {INCOMA Ltd.},\n\tauthor = {Beersmans, Marijke and de Graaf, Evelien and Van de Cruys, Tim and Fantoli, Margherita},\n\tyear = {2023},\n\tpages = {1--12},\n}\n\n\n\n","author_short":["Beersmans, M.","de Graaf, E.","Van de Cruys, T.","Fantoli, M."],"key":"beersmans_training_2023","id":"beersmans_training_2023","bibbaseid":"beersmans-degraaf-vandecruys-fantoli-trainingandevaluationofnamedentityrecognitionmodelsforclassicallatin-2023","role":"author","urls":{"Paper":"https://aclanthology.org/2023.alp-1.1.pdf"},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero-group/schulzkx/5158478","dataSources":["JFDnASMkoQCjjGL8E"],"keywords":[],"search_terms":["training","evaluation","named","entity","recognition","models","classical","latin","beersmans","de graaf","van de cruys","fantoli"],"title":"Training and Evaluation of Named Entity Recognition Models for Classical Latin","year":2023}