Dr. Szafir's research bridges data science, information visualization, and visual cognition to model the kinds of information people intuitively extract from data visualizations. Through this process, she derives quantified insight into the role of perception in interpreting visualizations and other visual interfaces by gauging how real viewers in natural environments perceive encoded information. She uses these models to develop novel interactive systems that support more accurate data exploration at dramatically larger scales than previously possible and that couple expert analysis with automated methods like interactive machine learning. The resulting systems address research problems across a variety of domains, including earth science, defense, emergency response, biochemistry, and the humanities. Her work also extends these practices to novel interface technologies to generate a foundational understanding of effective application design grounded in human perception and cognition.
information visualization, perceptual computing, augmented reality, human-computer interaction, exploratory data analysis, visual cognition, perception, color science, crowdsourcing, digital humanities, bioinformatics