Keith Harrigian is a Machine Learning Scientist at Netflix, where he works on content-related forecasting. He previously earned his PhD in Computer Science from Johns Hopkins University. With his advisor Mark Dredze, Keith researched computational and statistical methods for modeling natural language in the context of promoting health equity. You can learn more about his past and ongoing work here.

Recent News

July 2024 - I successfully defended my PhD Thesis “Towards Robust Natural Language Processing to Promote Health Equity.” Slides from my defense can be found here. My full thesis document can be found here.

February 2024 - I gave a talk at the Center for Language and Speech Processing (CLSP) about using AI to fight inequity in healthcare. Slides available here. Recording available here.

November 2023 - New paper alert: “An Eye on Clinical BERT: Investigating Language Model Generalization for Diabetic Eye Disease Phenotyping” will be presented at ML4H in New Orleans next month. Our study tempers recent claims that language models pretrained on clinical data are necessary for clinical NLP tasks and highlights the importance of not treating clinical language data as a single homogeneous domain. Find the paper here.

October 2023 - I was invited to speak with the Alzheimer’s Association’s AI working group about opportunities and challenges for AI in the fight against healthcare disparities. Slides available here.

July 2023 - New paper alert: “Characterization of Stigmatizing Language in Medical Records” has been accepted to appear in the main proceedings of ACL 2023. We examine what makes harmful language in clinical documentation different from other forms of abusive language in more widely-studied domains. Paper can be found here. Code, data, and models can be found here.