I'm currently working as a postdoc for the Ubiquitous Knowledge Processing (UKP) lab at TU Darmstadt. My dissertation research was primarily in automatic metaphor detection, semantic processing, and machine learning. I am fascinated by the variety of language utterances people produce and understand, and my work tends to deal with the computational frameworks necessary to handle these novelties. My dissertation research is on how to better represent syntactic information for automatic detection and interpretation of metaphors. I've employed novel methods for improving training data based on linguistic analysis, and improved the performance of state-of-the-art systems for multiple metaphor tasks
I have also worked extensively with professor Leysia Palen in the Department of Computer Science, applying NLP technology to the understanding of social media data produced during natural disasters. We worked with the National Center for Atmospheric Research to use modern computational methods to understand social media responses and reactions to natural disasters. This work culminated in the production of annotation schema and machine learning techniques, employing NLP methods to better identify vulnerable populations, assess their risk, and analyze their decision making behavior.
For more information on the projects I've worked on, check out my projects page!