My research concerns provenance, data science, and visualization. As data continues to proliferate, I am interested in methods that improve our ability to understand and use it. Algorithms help to summarize, filter, and mine data, and visualization techniques provide intuitive and interactive methods for examining data. Both are important ingredients, but keeping track of all these computations and explorations places a burden on users. Automated capture of provenance--the record of how results are achieved--provides users greater freedom to investigate data without manually recording every detail of each step.
I serve as one the core developers of the VisTrails project which is built with provenance as a central component. My dissertation research at the SCI Institute of the University of Utah focused on how provenance can be used to accelerate scientific discovery and improve usability as well as data management techniques to facilitate these uses. Provenance is generally noted for its use in verifying analyses and reproducing past work, but it can also be used, for example, to automate repetitive tasks or suggest new computations. Throughout my research, including time at Utah and NYU, I have had the opportunity to work with a variety of people and projects that span a number of fields, including the UV-CDAT climate project, the USGS-Fort Collins SAHM project, the DataONE project, the ALPS project, and the Cornell eBird project.
Current projects include: