University of Minnesota
University Relations
http://www.umn.edu/urelate
612-624-6868

Minnesota Supercomputing Institute


Log out of MyMSI

Research Abstracts Online
January 2008 - March 2009

University of Minnesota Twin Cities
College of Liberal Arts
School of Statistics

PI: Xiaotong Shen

Large-margin Hierarchical Classification

Hierarchical classification is critical to knowledge and context management as well as knowledge exploration, as in gene function classification and discovery and document categorization. In hierarchical classification, an input is classified by a structured hierarchy. In this project, the researchers are developing large-margin methods for hierarchical classification, with efforts focused toward gene function discovery. Large-margin methods involve support vector machines and psi-learning, which are designed to take the inter-class relationship into account for hierarchical classification. Some computational tools will be developed for large-scale problems, particularly for gene function discovery where gene functions are organized in a hierarchical fashion with high levels containing general information and lower levels giving specific information.

Group Member

Junhui Wang, Department of Statistics, Columbia University, New York, New York