How we perceive history is intricately woven with the narratives we inherit and the stories we are told. Yet, beneath the surface lie countless untold tales, particularly those of marginalised voices. In an era of technological advancement, artificial intelligence is emerging as a powerful tool to unveil these hidden histories, enriching our understanding of the past. VOICES, a groundbreaking project launched at Trinity College Dublin, spearheaded by an interdisciplinary team of historians and computer scientists, aims to shed light on the often-overlooked experiences of women amidst extreme trauma and civil unrest during early modern Ireland.
In this month’s episode of ADAPT Radio listeners had the privilege of hearing from two leading figures behind the VOICES project: Professor Jane Ohlmeyer, Erasmus Smith’s Professor of Modern History, and Professor Declan O’Sullivan, Professor in Computer Science in the School of Computer Science and Statistics.
During the podcast, Prof. Ohlmeyer highlighted the transformative role women played during the 16th and 17th centuries, both as promoters and resisters of colonisation and imperialism. However, she also acknowledged the daunting challenges researchers face when analysing historical data from this period, including inconsistencies in spelling, grammar, and euphemisms. To overcome these obstacles, Prof. O’Sullivan elaborated on the project’s innovative use of AI-powered tools called ‘transcribers’. These tools harness machine learning algorithms to decipher handwritten documents, converting them into analysable text. The journey begins with accessing raw data, followed by a meticulous process of identifying and connecting key information into a knowledge graph—a dynamic network akin to social network analysis.
Importantly, the digital archives generated by the VOICES project are intended to be shared with researchers, historians, and individuals with a vested interest in history. Prof. Ohlmeyer emphasised the importance of collaborative knowledge sharing within the historical community, highlighting the project’s commitment to open-source principles. As outlined by Prof. Ohlmeyer, by leveraging Ireland as a case study, the methodologies developed through VOICES are poised to be adaptable to various geographical and temporal contexts. This endeavour represents a significant milestone in historical research, marking the first systematic application of knowledge graphs in the humanities.
As the VOICES project progresses, it promises to amplify marginalised narratives, challenge existing historical paradigms, and pave the way for a more inclusive understanding of the past. With its dedication to transparency and open access, VOICES exemplifies the potential of AI-driven historical research to reshape our collective memory.
Learn more about the VOICES project here.
Listen here or wherever you get your podcasts.
ADAPT Radio: HumanAIse is ADAPT’s newest podcast series providing an in-depth look at the future of AI, automation and the implications of entrusting machines with our most sensitive information and decisions.
