Page Embeddings: Extracting and Classifying Historical Documents with Generic Vector Representations

Open Access
Authors
  • Carsten Schnober
  • Renate Smit
  • Manjusha Kuruppath
  • Kay Pepping
Publication date 2024
Host editors
  • Wouter Haverals
  • Marijn Koolen
  • Laure Thompson
Book title Proceedings of the Computational Humanities Research Conference 2024
Book subtitle Aarhus, Denmark, December 4-6, 2024
Series CEUR Workshop Proceedings
Event 2024 Computational Humanities Research Conference
Pages (from-to) 999-1011
Number of pages 13
Publisher Aachen: CEUR-WS
Organisations
  • Faculty of Humanities (FGw) - Amsterdam Institute for Humanities Research (AIHR) - Amsterdam School for Heritage, Memory and Material Culture (AHM)
Abstract We propose a neural network architecture designed to generate region and page embeddings for boundary detection and classification of documents within a large and heterogeneous historical archive. Our approach is versatile and can be applied to other tasks and datasets. This method enhances the accessibility of historical archives and promotes a more inclusive utilization of historical materials.
Document type Conference contribution
Language English
Published at
https://ceur-ws.org/Vol-3834/paper73.pdf (Final published version)
Other links
Downloads
paper73 (Final published version)
Permalink to this page
Back