Dr Aditi Dutta MSc, PhD

Dr Aditi Dutta

School of Government
Research Fellow in Computational Social Science

Contact details

Address
Muirhead Tower
University of Birmingham
Edgbaston
Birmingham
B15 2TT
UK

Aditi Dutta is a Research Fellow working at the intersection of computational social science, natural language processing, and climate and health policy. She is a member of Working Group 5 (Global team) of the Lancet Countdown project, developing indicators of public and political engagement with climate change.

Qualifications

  • PhD in Politics, University of Exeter, 2026
  • Enrichment Student, The Alan Turing Institute, 2024
  • Open Life Science (OLS) Fellow (cohort 7), 2023
  • Clinton Global Initiative Fellow, 2021
  • MSc in Data Science, University of Southampton, 2021
  • BTech in Computer Science and Engineering, Institute of Engineering and Management, Kolkata, 2019

Biography

Aditi Dutta is a Research Fellow in Computational Social Science in the School of Government at the University of Birmingham, where she works as part of Working Group 5 of the Lancet Countdown on Health and Climate Change. Her work focuses on tracking public and political engagement with health and climate change globally, using data science techniques including natural language processing, clustering, and statistical modelling to develop and refine global indicators contributing to the Lancet Countdown's annual report.

Aditi completed her PhD in Politics at the University of Exeter in 2026, where her doctoral research applied computational social science methods to the study of online sexism in political discourse. Her thesis explored ways to improve the reliability of language model annotations for subjective classification tasks, with a particular focus on quantifying model uncertainty. Alongside her doctoral research, she held postgraduate teaching and research assistant roles, teaching statistics to social science students and contributing to externally funded projects including the TWICEASGOOD project on media, sexism, and elections.

Prior to her doctoral studies, she completed an MSc in Data Science at the University of Southampton. She has also held research positions at GESIS — Leibniz Institute for the Social Sciences in Cologne, Germany, collaborating on projects examining data quality in digital social research.

Aditi is an affiliate of the Centre for Climate Change and Social Transformations (CAST) and collaborates with an international network of researchers working on climate change, public engagement, and digital inequality. She has presented her work at a range of international venues.

Research

Research interests

  • Computational Social Science
  • Natural Language Processing
  • Large Language Models
  • AI Annotation and Reliability
  • Model Uncertainty
  • Online Political Discourse
  • Public and Political Engagement
  • Climate and Health Policy
  • Global Indicators
  • Open and Reproducible Research

Current research

Aditi’s current research sits at the intersection of computational social science and climate and health policy. As part of the Lancet Countdown project, she develops global indicators tracking public and political engagement with climate change, working with large-scale policy datasets including official government and international data collections. Her methodological focus includes NLP, machine learning, and the development of frameworks to assess and improve the reliability of AI-driven research tools.

Previous research

Her doctoral research applied computational methods to the study of online political discourse, with a focus on sexism directed at female politicians on social media. This work developed novel prompt engineering techniques to improve language model performance on subjective classification tasks, and algorithms to quantify model uncertainty for ambiguous cases requiring human review.

Publications

Recent publications

Commissioned report

Sippy, T, Margetts, H, Modhvadia, R, Bruff, O, Perez, J, Dutta, A, Clery, E (ed.), Curtice, J (ed.) & Jessop, C (ed.) 2025, Artificial intelligence: How political orientation shapes attitudes to AI. British Social Attitudes Reports, no. 42, National Centre for Social Research. <https://natcen.ac.uk/publications/bsa-42-artificial-intelligence>

Conference contribution

Dutta, A 2025, Quantifying Climate Policy Action and Its Links to Development Outcomes: A Cross-National Data-Driven Analysis. in NeurIPS 2025 Workshop: Tackling Climate Change with Machine Learning., #27, NeurIPS, 39th Annual Conference on Neural Information Processing Systems, NeurIPS 2025
, San Diego, California, United States, 2/12/25. <https://www.climatechange.ai/papers/neurips2025/27>

Literature review

Dutta, A, Banducci, S & Camargo, CQ 2025, 'Divided by discipline? A systematic literature review on the quantification of online sexism and misogyny using a semi-automated approach', Scientometrics, vol. 130, no. 9, pp. 4915-4971. https://doi.org/10.1007/s11192-025-05410-2

View all publications in research portal