Die Maschine und die Geschichtswissenschaft : Der Einfluss von deep learning auf eine Disziplin

Hodel, Tobias (2022). Die Maschine und die Geschichtswissenschaft : Der Einfluss von deep learning auf eine Disziplin. In: Döring, Karoline Dominika; Haas, Stefan; König, Mareike; Wettlaufer, Jörg (eds.) Digital History: Konzepte, Methoden und Kritiken Digitaler Geschichtswissenschaft. Studies in Digital Historyand Hermeneutics: Vol. 6 (pp. 65-80). Berlin, Boston: DeGruyter 10.1515/9783110757101-004

[img]
Preview
Text
10.1515_9783110757101-004.pdf - Published Version
Available under License Creative Commons: Attribution (CC-BY).

Download (480kB) | Preview

Deep learning is a method from the field of artificial intelligence that is currently being used in many disciplines to create appraisal decisions. The form of machine learning is also being used in history, for example, for text recognition or the identification of named entities. Since deep learning will become a much stronger part of the methodological apparatus in the future, it is worth taking a critical look at what is happening. The moment of training plays a crucial role in the method. There, models are created and optimized. Based on the provided data, patterns can be recognized and imitated. Significantly, the created models are only verifiable in retrospect and with test procedures and are, at most, partially comprehensible. Thus, hermeneutic approaches are needed to understand and classify the models. Accordingly, the use of deep learning in history will entail a new reflection on methods, which must take into account technical circumstances on the one hand and disciplinary specifications on the other.

Item Type:

Book Section (Book Chapter)

Division/Institute:

06 Faculty of Humanities > Other Institutions > Walter Benjamin Kolleg (WBKolleg) > Digital Humanities

UniBE Contributor:

Hodel, Tobias Mathias

Subjects:

900 History

ISSN:

2629-4559

ISBN:

978-3-11-075710-1

Series:

Studies in Digital Historyand Hermeneutics

Publisher:

DeGruyter

Language:

German

Submitter:

Tobias Mathias Hodel

Date Deposited:

05 Sep 2022 15:21

Last Modified:

05 Sep 2022 15:30

Publisher DOI:

10.1515/9783110757101-004

BORIS DOI:

10.48350/172374

URI:

https://boris.unibe.ch/id/eprint/172374

Actions (login required)

Edit item Edit item
Provide Feedback