| Instructor |
Dr.
Gheith Abandah |
| Email |
abandah@ju.edu.jo |
| Home
page |
http://www.abandah.com/gheith |
| Office |
Computer
Engineering
405 |
| Office
hours |
 | Mon 11:00 - 12:00 |
| Tue 12:00 - 1:00 |
| Thu 10:00 - 11:00 |
|
| No.
of credit hrs |
3 |
| Prerequisites |
1901473: Operating Systems |
| Time
and room |
Mon and Wed 9:30-11:00,
CPE 001 |
|
Textbook |
Theodoridis S, Koutroumbas K (2006) Pattern recognition, 3rd edn.
Academic Press. |
| References |
Pattern Classification (2nd ed.) by
Richard O. Duda, Peter E. Hart and David G. Stork, Wiley Interscience, 2001. |
| Grading |
| Mid-Term Exam |
30% |
| Term Project |
20% |
| Final Exam |
50% |
| Policies |
|
Attendance is required. |
|
All
submitted work must be yours. |
|
Cheating
will not be tolerated. |
|
This
course requires significant effort. |
|
| Catalog Description |
Basic concepts in pattern
recognition. Classifiers, data mining, and knowledge discovery. Basic
concepts of decision functions. Linear decision functions, generalized
decision functions, and orthogonal functions. Classification by distance
functions and clustering. Minimum distance classification. Single
prototypes, multi-prototypes, and nearest-neighbor classification.
Clustering and clusters: threshold order-dependent clustering algorithm,
Max-Min distance method, c-means iterative algorithm (CMI). The ISODATA
algorithm. Classification using statistical approaches. A general Bayes
classifier. Normally distributed patterns: univariate, multivariate,
multiclass multivariate. Estimation of probability density functions.
Feature selection: introduction, distance measures, and clustering
transformations. Feature selection methods: entropy minimization, and
functional approximation. Fuzzy concepts: fuzzy set theory, the extension
principle, and fuzzy relations. Fuzzy and crisp classification. Fuzzy
clustering: fuzzy c-means iterative algorithm (FCMI), and fuzzy
partitioning. Fuzzy pattern recognition. Syntactic pattern recognition:
grammar types, selecting primitives, syntax analysis for recognition, and
stochastic languages. Introduction to NNs, the McCulloch-Pitts (MP) neuron,
Hebb NN, the Perceptron, the ADALINE, and Backpropagation NN and its
applications: Pattern classification using Neural Networks (NNs). |
| Tentative
outline |
|
Introduction |
|
Bayes Classifiers |
|
Linear Classifiers |
|
Non Linear Classifiers |
Midterm Exam
|
Feature Selection |
|
Feature Generation |
|
Template Matching |
|
Context Dependent Classification |
|
System Evaluation |
|
Clustering Algorithms |
Final
Exam
|
| Special Dates |
| Mon 8 Feb 2010 |
First Lecture |
| Mon 22 Mar 2010 |
Project Proposal Due |
| Wed 31 Mar 2010 |
Midterm Exam (9:30-11:00) |
| Mon 3 May 2010 |
Project Report Due |
| Wed 19 May 2010 |
Last Lecture |
| Sun 23 May 2010 |
Final Exam (12:00-2:00) |
|
| Handouts |
Slides
- Course
Introduction
- Introduction
- Bayes
Classifiers
- Linear
Classifiers
- Nonlinear
Classifiers
- Feature
Selection
- Feature
Generation
- Template
Matching
-
Context Dependent Classification
- System
Evaluation
- Clustering
Algorithms
Suggested problems:
Chapter 2: 2.2, 2.7, 2.8, 2.12, 2.31
Chapter 3: 3.4, 3.6, 3.9
(Some problems require programming)
Chapter 4: 4.1, 4.3
Matlab Pattern
Classification 1
Matlab Pattern
Classification 2
Feature Extraction
Feature Selection (Slides)
Matlab Pattern
Classification 3 The midterm exam was given on
Saturday 31/3/2010, Solution,
Problem 3 worksheet Grades
as of 18/5/2010 including all term work out of 50%. |
|
Important Links |
|