Skip to content
BY-NC-ND 4.0 license Open Access Published by De Gruyter Oldenbourg 2022

The Digital Humanities and the Ladino Press: Using Machine Learning to Extract and Analyze Visual Content in Historic Ladino Newspapers

From the book Jewish Studies in the Digital Age

  • Benjamin Charles Germain Lee

Abstract

La Vara, El Tiempo, and La Boz De Oriente represent three of the major historic Ladino newspapers published across the diasporic Sephardic Jewish world in the twentieth century. While Sephardic Jewish history and culture have received increasing scholarly attention in recent years, the vast corpus of Ladino newspapers largely remains unmined, and the field continues to be marginal from the perspective of Jewish Studies. In this chapter, I apply computational analysis of the visual content to explore the Ladino press at a macroscopic level. Using a machine learning model that I developed for my project, Newspaper Navigator, I have constructed a dataset of extracted photographs, illustrations, maps, comics, editorial cartoons, and advertisements from over 15,000 digitized pages of Ladino newspapers. This method represents an emerging approach to digital humanities research with periodicals and presents opportunities to facilitate access and research within Jewish Studies. With this extracted visual content, it is possible to study the transnational dynamics shaping Sephardic print culture and the broader Sephardic experience at an unprecedented scale. Accordingly, I describe my analysis of this visual content using emerging techniques in order to provide insights related to motifs and temporal trends. I offer this work as a case study in interdisciplinary research in the digital humanities and Jewish Studies. In addition, I offer methodologicalreflections related to applying emerging computational techniques to Jewish Studies. I conclude with a reflection on the ethical considerations of applying machine learning techniques to Ladino newspapers and, more generally, to Jewish cultural heritage.

© 2022 Walter de Gruyter GmbH, Berlin/Boston
Downloaded on 28.3.2024 from https://www.degruyter.com/document/doi/10.1515/9783110744828-010/html
Scroll to top button