B3 Non-refereed conference proceedings
Making Sense of Bureaucratic Documents : Named Entity Recognition for State Authority Archives (2024)

Poso, V., Lipsanen, M., Toivanen, I., & Välisalo, T. (2024). Making Sense of Bureaucratic Documents : Named Entity Recognition for State Authority Archives. In Archiving 2024 Final Program and Proceedings (pp. 6-10). Society for Imaging Science & Technology. Archiving, 21. https://doi.org/10.2352/issn.2168-3204.2024.21.1.2

JYU authors or editors

Publication details

All authors or editors: Poso, Venla; Lipsanen, Mikko; Toivanen, Ida; Välisalo, Tanja

Parent publication: Archiving 2024 Final Program and Proceedings

Place and date of conference: Washington, DC., USA, 9.-12.4.2024

eISBN: 978-0-89208-366-2

Journal or series: Archiving

ISSN: 2161-8798

eISSN: 2161-8798

Publication year: 2024

Publication date: 09/04/2024

Number in series: 21

Pages range: 6-10

Number of pages in the book: 106

Publisher: Society for Imaging Science & Technology

Publication country: United States

Publication language: English

DOI: https://doi.org/10.2352/issn.2168-3204.2024.21.1.2

Publication open access: Openly available

Publication channel open access: Open Access channel

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/97591

Abstract

The usability and accessibility of digitised archival data can be improved using deep learning solutions. In this paper, the authors present their work in developing a named entity recognition (NER) model for digitised archival data, specifically state authority documents. The entities for the model were chosen based on surveying different user groups. In addition to common entities, two new entities were created to identify businesses (FIBC) and archival documents (JON). The NER model was trained by fine-tuning an existing Finnish BERT model. The training data also included modern digitally born texts to achieve good performance with various types of inputs. The finished model performs fairly well with OCR-processed data, achieving an overall F1 score of 0.868, and particularly well with the new entities (F1 scores of 0.89 and 0.97 for JON and FIBC, respectively).

Keywords: digitising; optical character recognition; named-entity recognition; machine learning; archival materials; state archives

Contributing organizations

Related projects

Ministry reporting: Yes

VIRTA submission year: 2024