Viral Stereotypes: Analyzing Historical Circulations of Media Images through Deep Learning,
Pierre-Carl Langlais, Université Paul-Valéry Montpellier 3 and Sorbonne Université
You can see this video (enhanced version) in fullscreen here
Abstract
The digitization of large scale cultural heritage collections has opened up new perspectives for the study of global culture. Recent works in digital humanities have started to rely on automated tools to uncover viral networks of reprinted textual works. The Numapresse project aims to extend this approach to visual content, using a pilot corpus of 100,000 illustrations extracted from magazines and dailies from 1850 to 1914. The identification of reprinted images is based on a "twisted" deep learning model. Instead of using the classification output, we focus on an intermediary step: the data generated by the third layer of the neural network which encodes hybrid information between the original visual rendering and the abstract categories used for classification. This experimental tool has not only allowed us to retrieve hundreds of reprints but also to derive structural information on the "flow" of reprint from one periodic to another. For instance, it turned out that a large share of reuses occurred in the Voleur illustré (the illustrated thief), which acted as a nineteenth-century aggregator of visual content.
Biography
Pierre-Carl Langlais is a French researcher working on media studies and the digital humanities. His Ph.D. thesis on the birth of the financial column in the nineteenth-century press using distant reading methods was awarded the 2016 prize of the French society for information and communication studies. He is currently the post-doc of Numapresse, a digital humanities project dedicated to the exploration of massive archives of French digitized press. A long-time supporter of open access, Pierre-Carl Langlais wrote in 2016 a report on the emerging editorial model of open access for the French Ministry of research: this work inspired the Jussieu call for open science. In 2020 he published his first book in collaboration with Marie-Ève Thérenty and Julien Schuh at the publishing house of the CNRS, “Fake new & viralité avant Internet : les lapins du Père-Lachaise et autres légendes médiatiques.”
