Chris Jermaine - Research Statement
Overview. My work lies at the intersection of systems-oriented database research and applied probability and statistics, and is concerned with the design and implementation of innovative database systems software. I build large-scale, fully functional, prototype data management systems from the ground up. The systems I construct make use of probabilistic and statistical techniques to greatly expand the applicability of database software by making it run faster (for example, by using statistical inference to quickly “guess” the answer to a query) or by making it more flexible (for example, by providing a generic framework to handle missing, incomplete, or extrapolated data using tools such as Markov-Chain Monte Carlo methods). While the marriage of database system design and statistical analysis is quite natural—after all, statistics as a science is concerned with analyzing data in order to infer characteristics of the process that produced the data—the integration of statistical and data management software has been largely ignored over 35 years of database systems research.