Research project Natural Language Processing for Digital Art History
The recognition of the potential of computational tools and approaches for the study of cultural heritage has been growing in the last decades in different academic disciplines including art history.
The advancing digitization of artefacts and growing digital databases as well as the emergence of Open Access principles facilitate and boost these developments worldwide. However, scholars emphasize that the digital research of visual arts must be informed by professional art historical knowledge that acknowledges the situatedness of knowledge production (Bentkowska-Kafel, 2015). Paul Jascot (Jaskot, 2019) poses the question: “(. . . ) what are the critical questions in art history that demand and are best suited to specific digital methods?”
Taking this position as a conceptual point of departure for our research, our project combines ‘distant viewing’ (Arnold and Tilton, 2019) or macroanalysis of visual materials through Data Science and Natural Language Processing, and ‘close reading’ of the artefacts through formal and contextual analysis (Gold and Klein, 2019) in order to study Japanese early modern woodblock prints, so-called ukiyo-e (‘pictures of the floating world’), produced between the seventeenth and mid-nineteenth century.
Project description
The project is embedded in ”Machine Learning for Digital Art History”, an interdisciplinary and international network established in 2018 by three scholars: Dr. Ewa Machotka (Stockholm University, Art History of Japan), Prof. Panagiotis Papapetrou (Stockholm University, Machine Learning), and Dr. John Pavlopoulos (Ca’Forscari University of Venice, Computer Science, Data Science & NLP). The network also includes Konstantina Liagkiou (Athens University of Economics and Business, MSc in Data Science) and Marita Chatzipanagiotou (Athens University of Economics and Business, MSc in Digital Methods for the Humanities).
Project members
Project managers
John Pavlopoulos
Affiliated Researcher
Members
Panagiotis Papapetrou
Professor, deputy head of department
Konstantina Liagkou
Research Assistant
Marita Chatzipanagiotou
Research Assistant
Publications
More about this project
Project activities (2018-):
Peer-reviewed conference presentations:
Machotka E. and Pavlopoulos J. “Mapping meisho: NLP for Japanese Art History”, International Conference “The Digital Turn in Early Modern Japanese Studies”, December 2022, University of Cambridge.
Machotka E. “Mapping Japanese Landscape Prints with NLP: Challenges and Solutions”, International Workshop “Error Correcting HTR for Historical Manuscripts”, November 2022, Ca’Foscari University of Venice.
Pavlopoulos J. "HTR Error Correction: An Overview of the Recent HTREC Challenge", International Workshop “Error Correcting HTR for Historical Manuscripts”, November 2022, Ca’Foscari University of Venice.
Machotka, E. Pavlopoulos, J., Chatzipanagiotou, M., and Lagkiou K. ”Improving Named-Entity Recognition on titles of Ukiyo-e prints: Towards a ‘Distant Viewing’ in Art History”, Digital Humanities Conference 2022 (ADHO) Responding to Asian Diversity, July 2022, Tokyo.
Machotka E., Pavlopoulos J. Lagkiou K. “Distant Viewing of Ukiyo-e Prints.” The 13th Language Resources and Evaluation Conference (LREC), June 2022, Marseilles.
Machotka E., and Pavlopoulos J. “Automated Harvesting of Geospatial Data from ukiyo-e Prints with NLP.”, Annual Report Meeting, The International Joint Digital Archiving Center for Japanese Art and Culture, Ritsumeikan University, February 2022, Kyoto,
Machotka, E. Pavlopoulos, J., and Chatzipanagiotou, M. ”Automated Recognition of Geographical Named Entities in Titles of Ukiyo-e prints.” Digital Humanities Workshop (DHW 2021), December 2021, Kiev.
Organization of international conferences/workshops:
International Workshop “Error Correcting HTR for Historical Manuscripts”, November 2022, Ca’Foscari University of Venice. Machotka E. and Pavlopoulos J.: organization and chairing.
The International Conference Digital Humanities Now, Stockholm University (2022). Machotka E.: co-organization and chairing.
International Workshop Data Science for Digital Art History, at the KDD 2018 Conference on Knowledge Discovery and Data Mining, Londo, 2018 (peer-reviewed). Machotka E. and Papapetrou P.: organization and chairing.
Related activities:
ARC-iJAC Project Spotlights: An Interview on Natural Language Processing for a Geospatial Exploration of Ukiyo-e Prints & AI-powered Text Recognition of Inscriptions in Ukiyo-e Prints (Project Leader: Dr. Ewa Machotka, Stockholm Univ.)
All members of the project are also affiliated at the International Joint Digital Archiving Center for Japanese Art and Culture, Ritsumeikan University, Kyoto, Japan as Visiting Researchers.
Ewa Machotka acted as the Co-Director of the Doctoral School in Digital Humanities, Stockholm Univerity (2020-22).