Isabelle Lorge

I completed my PhD in Theoretical and Applied Linguistics at the University of Cambridge with Dr. Napoleon Katsos. My research focused on pragmatic (context-driven) inferences for word learning in bilingual and monolingual children using human data and Bayesian models. In April 2017, I interned at Microsoft Research Cambridge and collaborated with Alex Taylor from the Human Experience and Design (HXD) group and Nate Kushman from the Machine Intelligence and Perception (MIP) group on identification and understanding of context-specific features in dialog.

Between 2020 and 2022 I developed novel NLP models for financial applications at Arabesque AI with a team of PhD graduates from top universities. My research areas included sentiment analysis in news and social media, unsupervised learning and knowledge graphs. I then spent a year as Postdoctoral Research Assistant in Text Mining and Experimental Semantics in the Department of Engineering at the University of Oxford researching semantic composition and polarisation in social media with Professor Janet Pierrehumbert.

I am now a Research Scientist in the Department of Psychiatry at the University of Oxford working on the Chronosig project, which aims to use state-of-the-art NLP to improve mental health triaging in secondary care. As a psycholinguist and a pragmatician, I am interested both in how studying human behaviour can help us build better models and in how models can give us insight into human behaviour.