Publikation: GHisBERT – Training BERT from scratch for lexical semantic investigations across historical German language stages
Dateien
Datum
Autor:innen
Herausgeber:innen
ISSN der Zeitschrift
Electronic ISSN
ISBN
Bibliografische Daten
Verlag
Schriftenreihe
Auflagebezeichnung
DOI (zitierfähiger Link)
Internationale Patentnummer
Angaben zur Forschungsförderung
Projekt
Open Access-Veröffentlichung
Sammlungen
Core Facility der Universität Konstanz
Titel in einer weiteren Sprache
Publikationstyp
Publikationsstatus
Erschienen in
Zusammenfassung
While static embeddings have dominated computational approaches to lexical semantic change for quite some time, recent approaches try to leverage the contextualized embeddings generated by the language model BERT for identifying semantic shifts in historical texts. However, despite their usability for detecting changes in the more recent past, it remains unclear how well language models scale to investigations going back further in time, where the language differs substantially from the training data underlying the models. In this paper, we present GHisBERT, a BERT-based language model trained from scratch on historical data covering all attested stages of German (going back to Old High German, c. 750 CE). Given a lack of ground truth data for investigating lexical semantic change across historical German language stages, we evaluate our model via a lexical similarity analysis of ten stable concepts. We show that, in comparison with an unmodified and a fine-tuned German BERT-base model, our model performs best in terms of assessing inter-concept similarity as well as intra-concept similarity over time. This in turn argues for the necessity of pre-training historical language models from scratch when working with historical linguistic data.
Zusammenfassung in einer weiteren Sprache
Fachgebiet (DDC)
Schlagwörter
Konferenz
Rezension
Zitieren
ISO 690
BECK, Christin, Marisa KÖLLNER, 2023. GHisBERT – Training BERT from scratch for lexical semantic investigations across historical German language stages. 4th International Workshop on Computational Approaches to Historical Language Change 2023. Singapore, 6. Dez. 2023 - 10. Dez. 2023. In: TAHMASEBI, Nina, ed., Syrielle MONTARIOL, ed., Haim DUBOSSARSKY, ed. and others. Proceedings of the 4th Workshop on Computational Approaches to Historical Language Change. Stroudsburg, PA: Association for Computational Linguistics, 2023, pp. 33-45. Available under: doi: 10.18653/v1/2023.lchange-1.4BibTex
@inproceedings{Beck2023GHisB-69901, year={2023}, doi={10.18653/v1/2023.lchange-1.4}, title={GHisBERT – Training BERT from scratch for lexical semantic investigations across historical German language stages}, publisher={Association for Computational Linguistics}, address={Stroudsburg, PA}, booktitle={Proceedings of the 4th Workshop on Computational Approaches to Historical Language Change}, pages={33--45}, editor={Tahmasebi, Nina and Montariol, Syrielle and Dubossarsky, Haim}, author={Beck, Christin and Köllner, Marisa} }
RDF
<rdf:RDF xmlns:dcterms="http://purl.org/dc/terms/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:bibo="http://purl.org/ontology/bibo/" xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#" xmlns:foaf="http://xmlns.com/foaf/0.1/" xmlns:void="http://rdfs.org/ns/void#" xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/69901"> <dc:language>eng</dc:language> <dcterms:title>GHisBERT – Training BERT from scratch for lexical semantic investigations across historical German language stages</dcterms:title> <dc:contributor>Köllner, Marisa</dc:contributor> <dcterms:abstract>While static embeddings have dominated computational approaches to lexical semantic change for quite some time, recent approaches try to leverage the contextualized embeddings generated by the language model BERT for identifying semantic shifts in historical texts. However, despite their usability for detecting changes in the more recent past, it remains unclear how well language models scale to investigations going back further in time, where the language differs substantially from the training data underlying the models. In this paper, we present GHisBERT, a BERT-based language model trained from scratch on historical data covering all attested stages of German (going back to Old High German, c. 750 CE). Given a lack of ground truth data for investigating lexical semantic change across historical German language stages, we evaluate our model via a lexical similarity analysis of ten stable concepts. We show that, in comparison with an unmodified and a fine-tuned German BERT-base model, our model performs best in terms of assessing inter-concept similarity as well as intra-concept similarity over time. This in turn argues for the necessity of pre-training historical language models from scratch when working with historical linguistic data.</dcterms:abstract> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2024-05-02T10:33:43Z</dc:date> <dc:creator>Köllner, Marisa</dc:creator> <foaf:homepage rdf:resource="http://localhost:8080/"/> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/45"/> <dc:creator>Beck, Christin</dc:creator> <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2024-05-02T10:33:43Z</dcterms:available> <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/69901"/> <dcterms:issued>2023</dcterms:issued> <dc:contributor>Beck, Christin</dc:contributor> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/45"/> </rdf:Description> </rdf:RDF>