Claire Duiyi Dai is a computational social scientist and economist with cross-disciplinary training in economics, data science, and political communication. She specialises in the development and application of natural language processing (NLP), large language models (LLMs), and computer vision techniques to analyse political texts and media narratives at scale. She is particularly interested in using large-scale textual and visual data to understand policy-relevant phenomena across both traditional and emerging media environments.
Claire is currently a Researcher in Data Science and Informatics at the University of Oxford, based at the Leverhulme Centre for Demographic Science and the ESRC Centre for Care. Her projects include a large-scale analysis of UK parliamentary discourse on care using LLMs and speaker disambiguation techniques. She is also one of the developers of RobustiPy, an open-source Python library for multiverse and specification-curve analysis designed to improve robustness and transparency in empirical research.
She holds a PhD in Economics from the University of Birmingham, where her award-winning doctoral work combined unsupervised machine learning, computer vision, and macroeconomic modelling to understand Brexit-induced uncertainty and media bias. Beyond academia, Claire has contributed as an invited advisor to the Scottish Government, where she developed innovative text-based indicators of business sentiment to inform economic policymaking. She is also an experienced instructor and invited speaker on machine learning and textual analysis in the social sciences.
Duiyi Dai
Claire Duiyi Dai is a computational social scientist and economist with cross-disciplinary training in economics, data science, and political communication. She specialises in the development and application of natural language processing (NLP), large language models (LLMs), and computer vision techniques to analyse political texts and media narratives at scale. She is particularly interested in using large-scale textual and visual data to understand policy-relevant phenomena across both traditional and emerging media environments.
Claire is currently a Researcher in Data Science and Informatics at the University of Oxford, based at the Leverhulme Centre for Demographic Science and the ESRC Centre for Care. Her projects include a large-scale analysis of UK parliamentary discourse on care using LLMs and speaker disambiguation techniques. She is also one of the developers of RobustiPy, an open-source Python library for multiverse and specification-curve analysis designed to improve robustness and transparency in empirical research.
She holds a PhD in Economics from the University of Birmingham, where her award-winning doctoral work combined unsupervised machine learning, computer vision, and macroeconomic modelling to understand Brexit-induced uncertainty and media bias. Beyond academia, Claire has contributed as an invited advisor to the Scottish Government, where she developed innovative text-based indicators of business sentiment to inform economic policymaking. She is also an experienced instructor and invited speaker on machine learning and textual analysis in the social sciences.