Özlem Uzuner specializes in natural language processing, machine learning, and data science applications to real world problems. Her passion is to improve automatic language understanding, even when the narratives are written in ungrammatical, fragmented, and short-hand ambiguous utterances. She studies the language of electronic health records for improving information access and contributing to better healthcare. She develops natural language processing methods for understanding consumer-written text, such as electronic petitions describing citizen concerns and social media, for improving public policy. She studies sentiment and opinion analysis to uncover hints of deception and fraud in financial disclosure documents. Her research interests fall broadly under the umbrellas of health informatics, public health informatics, mental health informatics, and computational social science.
Uzuner not only develops novel methods for natural language understanding, she also develops and releases data sets that enable the advancement of the state of the art in natural language processing. Since 2006, she has developed numerous data sets in collaboration with Harvard Medical School, Brigham and Women’s Hospital, Massachusetts Institute of Technology, Veteran’s Administration at Salt Lake City, University of Cincinnati, University of Texas Health Sciences Center at Houston, and others. She serves on the editorial boards of several journals including the Journal of Biomedical Informatics and Journal of Intelligent Information Systems. She has guest edited several supplemental issues of the Journal of Biomedical Informatics and is the author of 70+ peer reviewed publications. Her work has been funded by the National Institutes of Health, National Library of Medicine, Office of the National Coordinator for Health Information Technology, and Philips Research NA.