Pattern Recognition

Spring 2006


last updated: 2006/03/29

1. Goal of Course

Pattern recognition can be broadly defined as a process to generate a meaningful description of data and deeper understanding of a problem through manipulation of a larger set of primitive raw data. More specifically we will be answering question: “What is this?


2. Recommended Textbook

Pattern classification, Second Edition, Duda, Hart and Stark


3. Instructor

[more details]


Dr. M Asmat Ullah Khan.  (Pakistan)

Contact Information

Tehephone : 031-201-2493
E-mail :
Office :

Electronic & Information Building #B08


4. Tentative Schedule (tentative)



Lecture 1

Introduction, Fundamental steps in Pattern recognition.

Lecture 2

Probability and Random Variables.

Lecture 3

Random Vectors, Co-variance matrices.

Lecture 4

Vector Spaces.

Lecture 5

Linear Transformations.

Lecture 6

Eigen-values and Eigen Vectors.

Lecture 7

Introduction to decision theory, Likelihood ratio, Bayes risk.

Lecture 8

MAP, MAL criterion, Discriminant Functions.

Lecture 9

Dimentionality reduction and PCA.

Lecture 10

LDA vs. PCA, LDA variants.

Lecture 11

Linear and quadratic classifier part1.

Lecture 12

Linear and quadratic classifier part2.

Lecture 13

The K-nearest neighbor rule.

Lecture 14

Density estimitaion.

Lecture 15

Validation part1.

Lecture 16

Validation part2.


Project Presentation

You can download here:


5. Grading Policy

There will be one long project during the semester for the course. The project will be evaluated by a written report and presentation.


6. Download the syllabus file (MS-WORD) :      [DOWN]