EEL6825: Syllabus

Instructor: John G. Harris
Office: NEB 453
Phone: (352) 392-2652
Email: harris@cnel.ufl.edu
Homepage: http://www.cnel.ufl.edu/analog/harris.html
Office hours: M period 3 (9:35am-10:25), W periods 3-4 (9:35am-11:30)
US Mailing Address:
Department of Electrical and Computer Engineering
University of Florida
453 NEB, Bldg. 33
PO Box 116130
Gainesville, FL 32611-6130

Prerequisite: The course assumes previous programming experience and some familiarity with linear algebra and statistics (normally obtained through undergraduate coursework).

Class Meeting: MWF 8th Period (3pm-3:50), CHE 237

Class Homepage: http://www.cnel.ufl.edu/analog/courses/EEL6825

Required textbook: Neural Networks for Pattern Recognition, Christopher M. Bishop, Oxford University Press, 1995.

Reference Books:

• Pattern Classification and Scene Analysis by Richard Duda and Peter Hart, John Wiley & Sons 1973.
• Introduction to Statistical Pattern Recognition by Keinosuke Fukunaga, Academic Press 1990.

Course Objectives: This class provides an introduction to classical pattern recognition. Pattern recognition is the assignment of a physical object or event to one of several prescribed categories. This semester will include sections on neural networks and provide links to the classical statistical pattern recognition techniques. Applications include automated speech recognition, fingerprint identification and optical character recognition. Major topics include:

• Bayesian decision theory.
• Parametric estimation and supervised learning.
• Linear discriminant functions.
• Nonparametric methods.
• Feature extraction for representation and classification.
• Neural networks for pattern recognition

Grade Determination:

• 25% Homework
• 25% Exam 1
• 25% Exam 2
• 25% Final Project (due at the end of the semester)

The final project will be discussed in more detail as the semester progresses.

Exam Dates:

• Exam 1, TBA
• Exam 2, TBA
• Final Exam, TBA