we are grateful to God that a paper arising from an Undergraduate Research Experience on Campus (URECA) student-project has been accepted to be read at Interdisciplinary Perspectives: Bridging Sociological Studies in the Digital Age, hosted by King’s College London, as well as identified as a Showcase at this year's Research Symposium of the Centre for Information Technology in Education, The University of Hong Kong (CITERS 2024). the former is organised by doctoral candidates of the Department of Digital Humanities at King's, and aims to foster discussions and collaborations across various disciplines, specifically including Natural Language Processing (NLP), Social Sciences, and Social Media and Digital Methods in broader research in Digital Humanities.
the paper is titled Deriving insights from historical data of Singapore with computational text analysis: a case example of the Digital Humanities.
my student Daniel K Handoyo will share his work on the 4th of May in Hong Kong and on the 11th of June in London, and its abstract reads:
This study adopts a hands-on approach to the exploration of digital humanities, applying natural language processing (NLP) techniques to enhance the analysis of Singapore's historical data. The sharing is relevant to the symposium theme in that it speaks towards how domain-specific epistemologies – of history, in the present case - intersect with literacies associated with Artificial Intelligence (AI) and Data Science. In this way, the sharing is relevant to the symposium topic of ‘digital literacy for future readiness’. We focus our inquiry on the recorded speeches and debates of Singapore’s early parliament, offering a window into the legislative priorities that shaped the nation’s policy landscape. In this study, we employ topic modeling – an unsupervised machine learning technique for discovering latent themes in a collection of documents, through the use of BERTopic, a cutting-edge framework that leverages transformer-based language models. This approach enables us to derive meaningful patterns and trends from parliamentary discourse, such as determining correlations between individual Members of Parliament (MPs) and the topics they frequently engaged with. This research illustrates how computational methods can be integrated with traditional historical analysis, underscoring the transformative potential of digital tools in supporting scholarly research within the humanities.