Course objectives
The goal of this course is to provide students with understanding of
the fundamentals of statistical pattern recognition and the ability to
implement key methods in MATLAB.
Students will write custom MATLAB functions to perform many pattern
recognition tasks and we will also draw upon some existing tools available
in the public domain.
Grading
The grade will be computed as follows:
|
Homework:
|
10%
|
|
Projects:
|
40%
|
|
Mid-term exam:
|
25%
|
|
Final exam:
|
25%
|
Prerequisites
A prerequisite for ECE661 is listed on the books. However, students with knowledge of
probabilities (ECE503), linear algebra basics, and a working knowledge of
MATLAB should be prepared.
MATLAB
The homework, projects and many in-class assignments will be done using
the MATLAB environment. MATLAB with
several useful toolboxes is available in the School of Engineering
in KL272. The student version
of MATLAB is available for Windows and Macintosh (published by Prentice
Hall) through the UD Computer
Store for about $100. The
statistics toolbox has a number of relevant functions and I would recommend
getting it for this course (but it is not necessary).
A tutorial to
introduce the basic syntax and use can be found at the Mathworks web site: Getting
started with MATLAB. The website
for my ECE203 MATLAB
course may be helpful as well.
Here is an old but still useful primer.
Homework
Homework will be announced in class and posted on WebCT. Each problem will be graded with a 0, 1, or
2 (2 complete and correct, 1 effort shown, 0 little or no effort
shown).
Projects
Several MATLAB projects will be assigned during the semester. These will be announced in class and
posted on WebCT.
Exams
One midterm exam will be given during the semester and the date will be
announced in class. Tests will cover
material discussed in lectures and in specified portions of the text.
The final exam will be held on Monday,
December 10, 4:30 PM - 6:20 PM in KL 272.
Student’s Responsibility
It is the student’s responsibility to get any information covered
in class and posted on this website or WebCT so please check these
frequently (especially if you miss a class). Some lectures will treat material not
covered in the text. Participation
in class is strongly encouraged.
Feel free to ask questions.
Finally, enjoy the class!!
Tentative Course Outline
Chapter 1:
Introduction
Chapter 2:
Random Vectors
Chapter 3:
Hypothesis Testing
Chapter 4:
Parametric Classifiers
Chapter 5:
Parameter Estimation (Selected Topics)
Chapter 6:
Nonparametric PDF Estimation (Pazen and kNN)
Chapter 7:
Nonparametric Classification (kNN)
Chapters 9
& 10: Feature Selection and Extraction (Selected Topics)
Chapter 11:
Clustering (Selected Topics)
Other
Topics: Support Vector Machines,
Neural Networks, Anomaly Detection
|
