Class web page: http://mil.ufl.edu/~nechyba/eel6825
Class meeting: T: 4th; Th: 4th & 5th, NEB 201.

Suggested textbook: Richard O. Duda, Peter E. Hart and David G. Stork, Pattern Classification, 2nd ed. , John Wiley & Sons, New York, 2001 (ISBN 0-471-05669-3). Additional readings, including lecture notes, slides and selected papers from the literature will be posted periodically on the class web site.

Prerequisites:

• Basic linear algebra and statistics
• Previous programming experience (e.g. C/C++, MATLAB, Mathematica, etc.)

On-line stuff:

• Course materials updated 10/21
• Assignments updated 10/21
• Source code updated 09/08

This class introduces methods in pattern recognition and classification. Major topics include:

• Bayesian decision theory
• Parametric estimation and supervised learning
• Unsupervised learning and clustering
• Linear discriminant functions
• Neural networks
• Nonparametric methods
• Applications

The basic goal of this course is for you to (1) know and understand the most common pattern recognition algorithms in use today; and (2) understand when and how these algorithms can be applied successfully to particular applications. Throughout, we will reinforce theoretical discussions with real-world examples in areas like computer vision and speech recognition.

• 5%: Quiz #1 (09/18, in-class)
• 25%: Midterm (date/format TBA)
• 30%: Homework/mini-projects
• 40%: Semester project

Many class announcements, clarifications and answers to student questions will be distributed primarily via e-mail. To get on the class e-mail list, you should send an e-mail to nechyba@mil.ufl.edu with the subject of the e-mail being EEL6825; be sure to include your full name in the body of the e-mail.

Some homeworks/mini-projects and the final project will require the use of a mathematical software package, such as Mathematica, MATLAB, MathCad and Maple, all available at student prices; which software package you choose is entirely up to you. Having said that, however, Mathematica will be used almost exclusively for in-class demonstrations and examples, as it is the most sophisticated general purpose mathematical software package available, allowing text, equations, graphics, numerical and symbolic mathematics to be seamlessly integrated into notebooks. These notebooks will be distributed on-line, and can be viewed with free software (MathReader) available from Wolfram, Inc., the developer of Mathematica. To modify and experiment with the Mathematica notebooks, however, will require that you have access to a copy of Mathematica.