报告摘要：Recently many matrix-based algorithms have been developed for
solving various data mining problems. In this talk, I will describe our
recent research efforts in the area. I will start with an application of
mining system log data to motivate clustering algorithms based on matrix
approximations. I will then describe several recent advances on Non-negative
Matrix Factorizations (NMF) for clustering. Finally I will discuss our
recent work on extending NMF to solve other data mining problems include
consensus-clustering, semi-supervised clustering, and feature selection.

At the end of talk, I will introduce and discuss the opportunities of the
graduate studies at the School of Computing and Information Sciences (SCIS)
at the Florida International University (FIU), a public research university.

报告人简介：Dr. Tao Li is currently an assistant professor in the School of
Computing and Information Sciences at Florida International University. He
received his Ph.D. in computer science from the Department of Computer
Science, University of Rochester in July 2004.  His research interests are
in data mining, machine learning, information retrieval, and bioinformatics.
He is the recipient of NSF CAREER Award (2006-2011) and IBM Faculty Research
Awards (2005 & 2007). He has published prolifically in top journals and
conferences and has severed extensively on the program committees of many
international conferences. More information about him can be found at
http://www.cs.fiu.edu/~taoli.