Design of Experiments and Data Mining
Department of Applied Mathematics
Illinois Institute of Technology
Stuart Building 113
Wednesday, March 28th, 2012
Abstract: Design of experiment (DOE) has been an important field in statistics. It contains many classic methodologies and theories. But due to the advanced development of modern computing technology and the invention of many new statistical fields, new techniques, methods, and theories in DOE are also introduced. For example, the birth of computer simulation, or in other words "computer experiments", lead to the development of a completely new branch in DOE, i.e., design of computer experiments. Meanwhile, the applications of DOE have become much broader than ever. In this talk, we will briefly review the important concepts and methodologies in DOE. Then we connect DOE to machine learning, and show how DOE techniques can help and boost some the machine learning methods.
Bio: Lulu Kang is an Assistant Professor at the Department of Applied Mathematics of Illinois Institute of Technology. She obtained her M.S. in Operations Research and Ph.D. in Industrial Engineering from H. Milton Stewart School of Industrial and System Engineering at Georgia Institute of Technology. She has done research in several fields in Statistics, including design and analysis of experiments, design of computer experiments, multivariate interpolation methods, meta-modeling methods, engineering statistics, etc. Currently she is focusing on Uncertainty Quantification field which is a new inter-discipline combining engineering, statistics and applied mathematics.