Digital resources in the Social Sciences and Humanities OpenEdition Our platforms OpenEdition Books OpenEdition Journals Hypotheses Calenda Libraries OpenEdition Freemium Follow us

CONFERENCE PAPER: “Where Did It Come From? Deep Learning for Event Extraction in Art Provenance”, 10/06/2022 – 10/07/2022.

From October 67th, 2022, Lab member Fabio participated in the 10th annual “New Directions in Analyzing Text as Data” conference (TADA), which is a leading forum for research on the study of politics, society, and culture through computational analysis of documents. At this year’s conference, Fabio presented a talk titled “Where Did It Come From? Deep Learning for Event Extraction in Art Provenance”. 

In the talk, he explored the possibilities offered by artificial intelligence in the structuring of provenance events, where events are composed of parties, locations, dates, and methods of transfer. In particular, Fabio focused on the possibilities of applying deep learning methods, i.e. tools using artificial intelligence, for event extraction from provenance. From a computer science perspective, the extraction of knowledge from a provenance text relates to an event extraction problem in Natural Language Processing (NLP), an area of research where computer science, linguistics, and, more recently, artificial intelligence intersect. Using a museum dataset of over 11,000 provenances, Fabio showed how it is possible to solve NLP tasks by training artificial intelligence. 

CONFERENCE PRESENTATION: “Probably Sold to Paalen, Possibly by Exchange”, 06/09/2022.

On June 9th and 10th, 2022, the University of Lausanne (UNIL) hosted the second edition of the Workshop on Computational Methods in the Humanities (COMHUM 2022). Fabio participated in the event with a presentation entitled “Probably Sold to Paalen, Possibly by Exchange: Vagueness, Incompleteness, Subjectivity, and Uncertainty in Digital Art Provenance”. Taking the example of the provenance of Picasso’s  ‘Head of a Woman’ (1927) now preserved at the MET, the talk focused on the challenges of, as well as modeling solutions for, VISU information (Vagueness, Incompleteness, Subjectivity, and Uncertainty). 

VISU information—from the Latin ‘de visu’ meaning “with your own eyes”—represents the effort and difficulty in reconstructing the provenance of a work of art. In digitizing and publishing art provenance as linked open data, Fabio argues, it is important that we include and model the complexity of VISU information. In so doing, we can avoid making historical records clear, complete, objective and certain, when they are simply not.

TEACHING: “Old Knowledge, New Techniques – Understanding Museum Data”, Winter Semester 2021/2022.

In the winter semester 2021/2022, Lynn and Fabio co-taught the seminar “Old Knowledge, New Techniques – Understanding Museum Data” as part of the Media Cultures module of the Digital Media Bachelor, a joint degree of Leuphana University and Hamburg Media School. The teaching proposes an experimental hybrid learning of provenance studies with digital knowledge representation.

“In the brave new electronic world, few bastions of the humanities have withstood the march of technology more tenaciously than the art museum.”

—Everett Elin, 1969

With the first Executive Director of the Museum Computer Network, Everett Elin’s, observation in mind, students were provided with theoretical and practical insights to critically analyze the current digital practices of museums in publishing art provenance online. On the one hand, the course presented the traditional methodology of provenance studies, providing literature and case studies. On the other, with continuous cross-disciplinary references, students learned the techniques and logic necessary to publish structured and machine-readable knowledge. These notions thus allowed students to practice what they had learned, experiencing the potential and the limits of the multidisciplinary approach. Working in groups, the students produced new and unique application scenarios of digital art provenance. Below you will find some examples of their work.

Fundamentally, the seminar was an opportunity to experiment on an educational level with what is already taking place at the Provenance Lab. It was motivated by the belief that academic training with a hybrid background is essential for providing institutions with adequate museum professionals who can meet digital challenges head-on.

AUTHORS: Nele Husen, Melanie Neuer, Hannah Prasuhn and Luise Pohlmann.

CONCEPT: “On a large screen in the museum there is a map of the world and a timeline, on which colourful dots represent different artworks. On Tablets or smartphone visitors can choose paintings, events and people they would like to learn more about.”

AUTHORS: Lelia van Hees, Isobel Hingston, Viktoria Dorothea Kamps and Nemanja Pulja.

CONCEPT: “Provenance data is displayed on a map paired with timestamps, making visible the space and time dimensions of a painting, as well as its possible network dimensions. This map is both accessible on the museum website and by way of a QR code in the museum itself .”

AUTHORS: Rebekka Skjørestad Halter, Marek Jakob Gollan, Charlotte Albert and Felizitas Klara Sophie Bader.

CONCEPT: “An interactive exhibition game in which school children go on a journey through time and space to track down a painting, collecting provenance data along the way. A trained educator guides them through the game, filling in any gaps and providing further context.”

NEWS STORY: “It is Time for Catalogues Raisonnés to Join the Digital Age”, 12/01/2022.

On December 1st, 2020, ahead of the International Catalogue Raisonné Association’s (ICRA) conference, Lynn talked to The Art Newspaper‘s Georgina Adam about how, “until recently, valuable information harvested through time-consuming and resource-intensive provenance research was often recorded in forms that could be understood by human beings but not by computers”.

You can read the article here.

INTERVIEW: “One Must Record the Inaccuracy”, 02/04/2020.

On February 24th, 2020, Lynn was interviewed by the national German weekly newspaper Die Zeit about how computer-based methods have thus far not been used for provenance research, but how her project at Leuphana University Lüneburg means to change. Amongst human biases, recording inaccuracies and the meaning of “messy or fuzzy data”, they talked about the “groundbreaking as well as promising” nature of Lynn’s project, which hopes to gain new insights into the distribution of artworks and thus make provenance research much more efficient.

You can read the article here.

NEWS STORY: “Lynn Rother Brings Art Research Forward”, 01/08/2020.

On January 8th, 2020, the German daily newspaper Die Tageszeitung ran an article on Lynn‘s appointment to the prestigious Licthenberg-Professorship, which is described as “the first chair on provenance research methods that exists in Germany”. In her position, the TAZ reports, Lynn hopes to “advance the humanities through computer-assisted methods”.

You can read the article here.