Download Selecting Data Mining and Statistical Procedures in Scientific

yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Nonlinear dimensionality reduction wikipedia , lookup

Department of Information and Systems Management
School of Business and Management
Hong Kong University of Science and Technology
Seminar Announcement
Seelleeccttiinngg D
Daattaa M
Miinniinngg aanndd S
Sttaattiissttiiccaall PPrroocceedduurreess
iinn S
Scciieennttiiffiicc R
Dr. Yonghong "Jade" Xu
The University of Memphis
22 June 2006 (Thursday)
11:00 am – 12:30 pm
ISMT Conference Room 4379 (L17/18)
 All interested are welcome 
Traditional statistical methods have limitations when used to analyze large-scale highdimensional data in business and scientific research. As an alternative, various new
procedures in "data mining and knowledge discovery" have been developed to search
for consistent patterns and/or systematic relationships between variables amidst large
amounts of data. However, selecting the appropriate procedures to fit the needs of
given research objectives is no trivial task. In this seminar, she will 1) highlight the
concepts of data mining including feature selection, model building, and pattern
definition, 2) compare data mining with multivariate statistical procedures in largescale data analysis and identify their strengths and weaknesses, 3) discuss the
selection of appropriate methods in various research scenarios, and 4) present a
research project in which multiple regression and Bayesian Belief Network are
compared to illustrate the similarities and differences of traditional statistical methods
and data mining procedures.
Dr. Yonghong "Jade" Xu is an Assistant Professor at the University of Memphis. She
specializes in educational statistics, large scale data analysis, and quantitative research
methodology. Dr. Xu has published several articles that explore both data mining and
traditional statistical techniques. Her current research interests include comparisons
of multivariate techniques, effective analysis of large national databases, and
examining the work life quality of faculty in postsecondary education.