Neil Marchant, recent PhD alum of the University of Melbourne in the School of Computing and Information Systems and former visiting scholar at Duke University in the Department of Statistical Science (jointly advised by Ben Rubinstein and Rebecca Steorts) has been named a Finalist for the Savage Award (Applied Methodology) for his thesis entitled “Statistical Approaches to Entity Resolution Under Uncertainty.”
In his thesis, he tackles a timely and important problem on entity resolution, which seeks to remove duplications across and within databases. His work is motivated by the Australian and United States Census, among other applications, where he proposes one of the first Bayesian frameworks that scales beyond millions of records, while quantifying uncertainty of the entity resolution task exactly, contrasting the previous literature. The code from his proposed work is available in open source and was run in house at the United States Census Bureau.
Neil is now a postdoctoral researcher in the Artificial Intelligence group at the University of Melbourne. His main research interests include machine learning and databases, including data integration, human-in-the-loop systems, and adversarial machine learning. He is a proponent of reproducible research and open-source code. Neil’s work has been supported by the National Science Foundation, the Sloan Foundation, and Australian Bureau of Statistics, the Australian Research Council, Australia's Defense Next Generation Technology Fund, and a DISER AUSMURI.