Bio
Aditi Dutta is a doctoral researcher at the University of Exeter, in the department of Politics. She is also a Postdoctoral Research Associate at the TWICEASGOOD project. Before starting her PhD, she had obtained her Master's in Data Science from the University of Southampton and was a Clinton Global Initiative (CGI) Fellow in the cohort of 2021. She is passionate about countering online sexism and other forms of online hate speech, to promote a safer digital space for all. Her current research focuses on understanding the dynamics of online sexism, where she uses social science theories to explain and improve computational models, using natural language processing. Aditi's research seeks to bring gender equality in digital spaces and investigate the different forms of sexism (and misogyny) that are propagated online, the subtle and implicit categories that go undetected by the models, eventually bringing about an impact in digital society research.
Research interests
Aditi's research relates to two of the Turing’s research themes: Online Safety and Ethics & Responsible Innovation. At The Alan Turing Institute, her research would focus on enhancing the identification of different forms of online sexism by integrating social linguistic scales. While sexism has been captured extensively in social science research, computational interpretation of such hate speech has been impacted due to biases, over-dependence on keywords, and other factors; essentially limiting the extent of computational models in replicating real-world scenarios. Aditi is interested in studying the linguistic characterisation of online sexism to capture its benevolent forms, while mitigating bias in the models. To address that, Aditi will work to develop a theoretical framework based on social science theories to capture the differences in types of online sexism, and improve the identification task.