Title: Non-parametric clustering
Cluster analysis, or clustering, is afundamental data mining task and diverse tool for big data analytics with vastapplications. During a clustering process, the user is typically required toprovide input such as specification of parameters to guide the search for thetarget clustering. However, in many cases the user has little to none knowledgeabout the required clustering parameters and is forced to enter uncertaininformation, leading to unfavorable clustering results. Thus it is important tostudy ``non-parametric" clustering that minimizes input uncertainty andrequires as few unjustified assumptions as possible.
Byron Gao is an associate professor ofcomputer science at Texas State University. He joined Texas State in 2008, wasa postdoctoral fellow at the University of Wisconsin in 2007-2008, and receivedPh.D. in 2007 and B.Sc. in 2003 from Simon Fraser University. His researchspans several related fields of data mining, databases, information retrievaland bioinformatics.