Industrial Mathematics Seminar Spring 2003:
A Pattern Analytic Approach to the Investigation of Data Sets in Industry
M676 Section-002 302151,  TR 2:10-3:25

Class meets in Weber 202 Tuesdays and Weber 205 Thursdays


Secure Class Web Page

This seminar will be conducted in collaboration with Siemens Corporate Research (SCR) is located in Princeton, NJ. SCR (http://www.scr.siemens.com/) provides Siemens operating companies with state-of-the-art imaging, software, multimedia and data analysis technologies. As such, Siemens is concerned with developing and applying cutting edge techniques in multivariate signal processing, pattern recognition and learning algorithms. During the course of this seminar students will investigate "real-world" data sets provided by SCR that will both illustrate elegant applications of existing abstract mathematical algorithms and also demonstrate the constant necessity for customization and
extension.

The topics we will consider for the evaluating of the Siemens data sets will include

The class will emphasize basic mathematical theory as well as algorithm development and analysis including algorithms available in the patent literature.  We will not survey these methods but restrict our attention to methods of current interest to SCR and our applications will be to Siemens data sets.

Coursework and Assessment:  projects will be group orientedand results will be evaluated by the instructor and Siemens. Students using this seminar in partial fulfillment of the MS program in Applied and Computational Mathematics will be required to present a summary of their projects to their Masters committee.

Prerequisites: Knowledge of a programming language, e.g., MATLAB, C++ or Java.  Linear Algebra (M369 or M560), Multivariable Calculus (M261)

Recommended Reading:  Geometric Data Analysis, M. Kirby;  The Elements of Statistical Learning: Data Mining, Inference, and Prediction by T. Hastie, R. Tibshirani, J. H. Friedman; Pattern Classification by Richard O. Duda, Peter E. Hart, David G. Stork.

Instructor: Professor Kirby, Weber 115, kirby@math.colostate.edu