Advancing data-intensive research at Brown through a centralized Data Science Practice
The CIS Data Science Practice collaborates closely with several research labs and PIs at Brown to build informatics pipelines and integrated data resources, and to apply, evaluate, and improve computational methods. In this talk, I will present preliminary results from three of our collaborations to illustrate how we work across disciplinary boundaries in the social sciences, life sciences, and the humanities:
1) Our work with the Rhode Island Innovative Policy Lab brings together economists and data scientists to integrate and analyze data from government and industry to understand whether existing polices work, then to design and rigorously test new policies to improve outcomes at lower cost. Research results from the lab feed directly into policy decisions in Rhode Island and DC, and guide the design of randomized control trials to measure the impact of new policies.
2) Working with evolutionary biologist Casey Dunn, HIV clinician/researcher Rami Kantor, and biostatistician Joe Hogan, we are developing new methods for understanding the transmission of HIV based on gene sequencing data from HIV-infected patients, with the ultimate goal of better identifying infected-unaware or untreated individuals.
3) Working with historian Jo Guldi, we are building text-mining tools for understanding large collections of historical text. Our approach combines new computational techniques with targeted expert curation to understand the changing debate over property law in British Parliament across a hundred-year time period.
Mark Howison is Director of Data Science in Brown's central IT department, Computing & Information Services. He received an B.S. in Mathematics from Brown and an M.S. in Computer Science from UC Berkeley, and previously worked as a Computer Systems Engineer at Berkeley Lab and an Application Scientist at Brown's Center for Computational & Visualization.