1
Classification Toolbox
For use with MATLAB
®
DavidG.Stork
Elad Yom-Tov
User’s Guide
2
Introduction
The classification toolbox is a list of functions for supervised and
unsupervised classification algorithms. These algorithms help design categorization
methods for experimental as well as synthetic data, using convenient graphic user
interfaces. The graphic user interface is designed for two-dimensional, two-class
problems, but most of the algorithms can be used on higher dimensional data, and
some can be used on multi-class data. In addition, higher dimentional data can be
reduced to two dimensions using one of several feature selection algorithms.
This user guide explains how to use the toolbox, for the novice as well as the
experienced user. Since most of the functions are operated from within the graphic
user interface, only basic knowledge of Matlab is required. No theoretic background
about classification is given in this manual, and the user is referred to [1] for such a
background.
The toolbox is intended to accompany a book to be published during 2003,
and users are encouraged to send comments and suggestions to elad@ieee.org
.
3
History and acknowledgements
This toolbox started as a course assignment in Dr. Ron Meir’s graduate course,
Pattern Recognition. The foundation for the toolbox, as well as most of the basic
algorithms, were coded by Elad Yom-Tov and Hilit Serby. This resulted in a toolbox
with (about) 15 algorithms.
A year later, Igor Makienko and Victor Yosef coded the Voted Perceptron
algorithm. Some time later Elad Yom-Tov continued coding, making the toolbox in
it’s present form with over 100 different algorithms.
The toolbox is now a pre-publication version of a work tentatively titled
Computer Manual in MATLAB to Accompany Pattern Classification by David G.
Stork and Elad Yom-Tov to be published during 2003.
4
Copyrights etc.
The toolbox is available from this address:
http://tiger.technion.ac.il/~eladyt/Classification_toolbox.html
Correspondence: Elad Yom-Tov
Faculty of Electrical Engineering
Technion – Israel Institute of Technology
Haifa 32000
Israel
elad@ieee.org
This is a pre-publication version of a work tentatively titled Computer Manual
in MATLAB to Accompany Pattern Classification by David G. Stork and Elad Yom-
Tov to be privately distributed to students enrolled in academic courses during 2002.
NOT FOR GENERAL DISTRIBUTION.
Copyright 2002 John Wiley & Sons, Inc. All rights reserved.
Reproduction, adaptation or any further distribution of this material is
expressly prohibited. For further information or to request permission for other uses,
please contact the Permissions Department, John Wiley & Sons, Inc., 605 Third Ave.,
New York, N.Y. 10158-0012. Telephone: 212-850-6011. Facsimile: 212-850-6008.
By using the Classification toolbox you agree to the following licensing
terms:
NO WARRANTY
BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE
PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN
WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT
WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO
THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE
DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY
COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM, BE
LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL
DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH
HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
I will, of course, thank you for any bugs you report to me (elad@ieee.org), but
I will not be held responsible for any problems or difficulties you encounter because
of them.
5
Table of contents
INTRODUCTION ....................................................................................................................................2
HISTORY AND ACKNOWLEDGEMENTS........................................................................................3
COPYRIGHTS ETC. ...............................................................................................................................4
TABLE OF CONTENTS .........................................................................................................................5
INSTALLATION .....................................................................................................................................6
SINGLE ALGORITHMS: THE GRAPHIC USER INTERFACE.....................................................7
A
SIMPLE EXAMPLE.................................................................................................................................8
S
ECTIONS OF THE GRAPHIC USER INTERFACE ..........................................................................................9
Menu ...................................................................................................................................................9
Input file area ...................................................................................................................................11
Classification and preprocessing parameters area .........................................................................11
Error estimation method........................................................................................................................... 12
Preprocessing........................................................................................................................................... 12
Classification algorithms.......................................................................................................................... 14
Graphic area ....................................................................................................................................16
Parameteric distributions area ........................................................................................................17
Command buttons.............................................................................................................................17
System messages box ........................................................................................................................18
Error percentages box......................................................................................................................18
F
EATURE SELECTION .............................................................................................................................19
A
COMPREHENSIVE EXAMPLE ...............................................................................................................20
COMPARING THE PERFORMANCE OF SEVERAL ALGORITHMS ......................................22
AN EXAMPLE ...................................................................................................................................23
FILE STRUCTURES.............................................................................................................................24
LIST OF FUNCTIONS..........................................................................................................................25
REFERENCES .......................................................................................................................................29
A
DDING A NEW PREPROCESSING ALGORITHM .......................................................................................30
A
DDING A NEW CLASSIFICATION ALGORITHM.......................................................................................31
A
DDING A NEW FEATURE SELECTION ALGORITHM ................................................................................32
INDEX .....................................................................................................................................................33