Harnessing Generative AI to Support Exploration and Discovery in Archival Collections

28th June 13:20 – 13:40

Speaker: Prof Richard Marciano

Abstract: This talk explores the use of Large Language Models (LLM) to facilitate the analysis and visualization of newspaper advertisements from the Maryland State Archives related to the trading of enslaved people. The study focuses on the Domestic Traffic Ads (DTA) collection of the State of Maryland between 1824 and 1864, which exposes chattel slavery practices where buyers and sellers would interact to exchange and share human beings often for social and domestic benefit. This case study is part of a larger project to explore computational treatments to remember the Legacy of Slavery (CT-LoS), towards reasserting erased memory. Previous studies have included computational treatments for Manumissions, Certificates of Freedom, and Runaway Slave Ads. Our approach is mindful of the social and ethical concerns that arise from using LLMs and the sensitivities related to working with slavery data. We are exploring a chatbot we call “ChatLoS”, as a querying tool, fine-tuned to be contextually aware of the DTA dataset which demonstrates trustworthy quantitative results.

Bio: Richard Marciano is a professor in the College of Information at the University of Maryland. He is also an affiliate faculty member in Engineering and Computer Science. Prior to that, he was a Professor at the School of Information and Library Science at the University of North Carolina at Chapel Hill for 5 years. He was also a Research Scientist at the San Diego Supercomputer Center (SDSC) at the University of California San Diego (UCSD) for 13 years. His research interests center on computational archival science (CAS), digital curation, digital preservation, sustainable archives, cyberinfrastructure, and big data. He is also the 2017 recipient of the Emmett Leahy Award for “outstanding and sustained work in digital records and information management”. He holds degrees in Avionics and Electrical Engineering, a Master’s and Ph.D. in Computer Science from the University of Iowa, and conducted a postdoc in Computational Geography.