This National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) Federal Grant Program (CFDA 47.070) award of $200,000 to Tennessee State University will investigate computational frameworks for cross-domain transfer learning of tabular data. The two-year project aims to develop deep clustering solutions for unlabeled tabular data and enable knowledge transfer across heterogeneous data tables. The research will leverage advancements in data distribution modeling and statistical methods to create cluster-friendly deep feature representations, facilitating subsequent transfer learning. The frameworks will be evaluated on electronic health record data for patient risk stratification tasks. Tennessee State University will share new algorithms, publications, and train students as part of this effort to advance tabular data science research.
Generated 3/25/25, 3:08 AM