Hi there! I'm a PhD candidate in the Linguistics Department at Stanford University, advised by Dan Jurafsky. I'm a member of the Stanford NLP group. I publish under my full name, Dorottya Demszky.
My research focuses on building and deploying natural language processing methods to support equitable and student centered education. I combine machine learning with linguistic theory and input from practitioners to develop interpretable and scalable education measures. By deploying these measures, I aim to improve educational interventions, teachers' professional development and the equity of curricular materials.
I am invested in understanding how NLP can be put to the service of social good. I primarily focus on solving problems in education, but in the past I also worked on dialect feature recognition, emotion detection, and on using natural language processing to understand political issues, such as polarization and propaganda.
I am grateful to my incredible collaborators, mentors and mentees, who have made our projects possible and have made me a better person and researcher.
You can reach me at ddemszky [at] stanford [dot] edu.
- Our team won an HAI Seed Grant for our project titled "Civics Education for a Just and Sustainable Future".
- I led a workshop on using NLP to analyze textbooks as part of the Women in Data Science workshop series.
- Our paper on measuring teacher's uptake of student contributions was accepted by ACL 2021.
- I gave a talk at the Cornell NLP seminar on my recent work on measuring teachers' uptake of student contributions.
- Our paper on recognizing dialect features was accepted by NAACL 2021.
- I was named as a Harvard CRCS Rising Star in AI for Social Good.