I'm a Ph.D. student at Carnegie Mellon University, more specifically at the Language Technologies Institute in the School of Computer Science. My advisor is Carolyn Rosé. Contact me at firstname.lastname@example.org.
My research focuses on representing a deeper understanding of natural language computationally. In my work, I focus on conversation, either online or in person. Complex linguistic and social cues are necessary for people to communicate well, but these subtleties of language are difficult for computational systems to model. My work involves defining representations of the flow of information, the social roles and choices that speakers make in a dialogue, and then building systems based on text mining and machine learning that can automatically identify those constructs. Much of my work is rooted in sociolinguistics, especially work in systemic functional linguistic inspired by Jim Martin.
I also have a secondary role as the developer of LightSIDE, a toolkit for non-expert users to quickly and easily utilize text mining technology. This tool incorporates feature extraction, model building, error analysis, and automated analysis, labeling, and prediction of data. LightSIDE is built in a modular interface that allows easy development of novel ideas and methods for feature extraction or other aspects of the pipeline from unstructured data to highly accurate and predictive output. The goal of the system is to be straightforward for users (even those with minimal machine learning background) to build machine learned models, understand the behaviors of those models, and label large amounts of data with high accuracy. Applications of LightSIDE have been highlighted (and occasionally vilified) in numerous sources including NPR, CNN, New Scientist, Education Week, the New York Times, and so on.