Recovering Lost Lives: Machine Learning to Surface African Women in Trans-Atlantic Slave Records

Clement Tetteh *

Department of History, University of Ghana.
 
Review
International Journal of Science and Research Archive, 2022, 07(02), 899-911.
Article DOI: 10.30574/ijsra.2022.7.2.0314
Publication history: 
Received on 07 October 2022; revised on 19 November 2022; accepted on 28 November 2022
 
Abstract: 
The historical record of the trans-Atlantic slave trade remains profoundly incomplete, particularly regarding the lives of African women whose identities were disproportionately erased, anonymized, or collapsed into generic labor categories. Existing archival collections including ship manifests, plantation inventories, baptismal rolls, court testimonies, sale ledgers, and manumission documents encode gender, kinship, ethnicity, and geographic origin unevenly due to colonial recordkeeping practices. As a result, women’s labor roles, reproductive histories, and social networks often disappear into ambiguous descriptors such as “girl,” “wench,” “daughter,” or unnamed household dependents. This research project proposes a machine learning framework to recover African women’s identities and reconstruct relational networks across fragmented documentary sources. The method integrates optical character recognition (OCR) to digitize handwritten and degraded archival documents; natural language processing (NLP) models trained to detect gender-coded vocabulary, kinship relational markers, and African naming patterns; and probabilistic entity resolution to match individuals across dispersed archival collections. Rather than displacing human interpretation, the approach is designed to augment historical reasoning, acting as a recovery tool that flags overlooked individuals and generates new research leads. Collaborative integration with existing digital humanities infrastructures especially Slave Voyages and Freedom on the Move enables scale, interoperability, and standardized metadata exchange. This project contributes to ongoing efforts in reparative archival work, feminist historiography, and Black Atlantic studies by systematically addressing archival silences. By leveraging computational approaches to illuminate erased presences, it advances a historically grounded, ethically sensitive framework for restoring African women to the narrative of Atlantic world history.
 
Keywords: 
Machine learning; Archival recovery; African diaspora; Trans-Atlantic slavery; Gender analysis; Digital humanities
 
Full text article in PDF: 
 
Copyright information: 

Copyright © 2022 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0