Computer Science Department

University of Western Ontario

CS 434a/541a Pattern Recognition

Fall 2004

Course Description

This is an introductory course to the theory of pattern recognition. Pattern recognition is concerned with assigning an object (or "pattern") to one of the several pre-specified categories (or "classes"). This classification is usually performed by finding and utilising useful features of an object. Recently there has been an explosion in computing power and digitised multidimensional data that needs to be analysed. Pattern recognition is a general tool for analysing multidimensional data. It is used in diverse fields for tasks such as handwriting recognition, lipreading, geological analysis, medical data processing, data mining, information retrieval, human-computer interaction, and so on. In this course we will study basic concepts in the field. We will cover Bayesian decision theory, maximum likelihood estimation, nonparametric estimation, linear discriminant functions, support vector machines, neural networks, unsupervised learning and clustering.

Lecture Hours: Monday 11-1 and Wednesday 12-1
Location: Middlesex College 316


Instructor Information

Instructor: Olga Veksler
Office: Middlesex College 361
Office Hours: 1:00 - 3:00 noon, Wednesdays or by appointment
E-Mail: olga [[at]] csd.uwo.ca
Phone: UWO extension 81417


Prerequisite

Note: Students are responsible for ensuring that they have either the prerequisites for this course, or written special permission from their Dean to enrol in. If a student does not have the course prerequisites, and has not been granted a special permission to take the course by the department, it is his/her best interest to drop the course well before the end of the add/drop period. If a student is not eligible for a course, he/she may be removed from it at any time, and will receive no adjustment to his/her fees. These decisions can not be appealed. Lack of prerequisites may not be used as the basis of appeal.

Required Text

R.O. Duda, P.E. Hart, D.G. Stork. Pattern Classification . John Wiley and sons, second edition. The book will be put on reserve in the library.

Course Content

Course Website

The website for the course is http://www.csd.uwo.ca/faculty/olga/Courses//CS434a_541a//index.html . Lecture notes, assignments, and class information will be posted on this website. You are responsible for reading this information frequently.

Lecture Notes

Some (but not all) of the lectures may be given in Power Point. I will post those lectures before or shortly after the lecture on the course web site. Occasionally I will also post lecture notes on the web site.

E-Mail Contact

We will occasionally need to send e-mail messages to the whole class, or to students individually. E-mail will be sent to your GAUL e-mail address. Make sure that you read your e-mail on GAUL frequently.

Best Way to Contact me

If you have a question and need to contact me, best way to do so is after the class or during my office hours. You may to contact me by email, but do not expect an answer within 2 minutes. I may be able to answer quickly, but it may take me several days. For questions requiring detailed explanations (like "I didn't get that concept, can you go over it again?"), I will ask you to come to my office hour. If the scheduled office hours are not convenient, please make an appointment.

Assignments

There assignments will be given approximately bi-weekely, and will involve both theoretical and programming exercises.

Midterm

Midterm will be given tentatively on November 8. I will try to give a sample midterm with solutions a week before the actual midterm. A makeup for the midterm will be arranged only if it falls on a religious holiday or in case of medical or family emergency. In any of these cases, please take the supporting documentation to the office of the Dean of your faculty.

Final Project

For the final project the students will design and test a pattern classification system. The students may choose one of several systems proposed by the instructor or may follow their own idea. The students will also write up a project report, 2 to 5 pages. The report will NOT be judged for its length. Rather it must include the 3 essential components. First you must state the problem you will try to solve, then the approach you are going to take to solve it, and last your results and what you have learned from your results. The results can be negative (that is my approach did not work so well), if they are negative, you should try to explain possible reasons for failure. The proposals for the final project will be due on November 1, and the final project is due on December 8. The week of November 22-28 I will ask you to either write me a one page report on how the final project is going or make an appointment to see me to discuss how the project is going. For the final project, student may collaborate in groups of two, in which case the project is expected to be more extensive.

Grading Scheme

Computing Facilities

Each student will have access to an account on the Computer Science Department senior undergraduate computing facility, GAUL. In accepting the GAUL account, a student agrees to abide by the department's Rules of Ethical Conduct.

Note: After-hours access to certain Computer Science lab rooms is by student card. If a student card is lost, a replacement card will no longer open these lab rooms, and the student must bring the new card to the I/O counter. Likewise, if a student card ceases to provide access where it should, it should be brought the I/O counter as well. There, the operator will swipe the card, record the complaint and send the information to the Systems Group who will send notice when they have fixed the problem.