Dr. Giulia Taurino and Prof. David Smith (Northeastern University) present "Re-Aggregating the Archive: reconstructing urban narratives through ML-augmented classification of photo-journalism collections," Thursday, September 29th, 5 - 6 pm, in The Nexus.

Refreshments to follow.

From the presenters: "'Photo morgues' are collections of photographs from a newspaper’s past issues and news cycles. Archived, or else in the jargon 'sent to the morgue', these files remained hidden in the history of journalism, in the form of defunct and forgotten newspaper clippings. From local events to nation-wide social movements, photo-journalism archival material stretches back for over a century and constitutes a unique resource to document the geo-cultural evolution of urban areas over time. Building upon Northeastern University Library’s efforts to digitize both published and unpublished images from the Boston Globe photo morgue, Seeing Our Neighborhoods: Providing Public Access to the Boston Globe Photograph Collection seeks to bring one of these collections back to life, reinserting otherwise fragmented records in the evolutionary cycle of the city and its inhabitants. Tapping into two main datasets of digital images from the Globe archive showing Boston’s development plans and social activism throughout the years, this project reconstructs neighborhoods as primarily experiential and intersubjective spaces (Lefebvre 1974), defined by multiple, interconnected cultural histories. Rather than disaggregating images into separate geo-localized units, we use machine learning to create an augmented classification and cataloging system able to recover and enrich the metadata available, thus rendering each record in the photo-archive as a fluid text (Bryant 2002) that can be manipulated and reorganized into ever new aggregates."

-

Hosted by the MIT Libraries and Programs in Digital Humanities, the Digital Humanities Workshop provides a flexible, dynamic space for faculty and researchers to connect with each other about digital research and teaching topics. Our Fall 2022 events will focus on mass digitization projects and visual analysis.

All MIT community members are welcome. Non-MIT guests and guests without an MIT ID should RSVP to lewer@mit.edu.