File Exchange

image thumbnail

Introduction to Classification

version (353 KB) by Richard Willey
Files and code from Computational Statistics: Getting Started with Classification using MATLAB®


Updated 01 Sep 2016

View License

This code is provides a simple introduction to some of the Classification capabilities in Statistics Toolbox. Key techniques used include
Using normplot to see whether features are normally distributed.
Using coercoff to look for correlation between features.
Using cvpartion to separate data into a test set and training set.
Training Naive Bayes classifiers and ensembles of decision trees.
Using sequentialfs to simplify a model

Comments and Ratings (22)

Krishna V

Lucy Sinha

Is it working for any type of data set??

Pravina K

pooja patil

can I get Matlab code for fuzzy c classifier or any other classifier that can perform on a satellite image?

Not so well introduced, I think. My understanding is that the code is an example of the feature selection technique. You can refer to the following documentation link for more details of the process:

The functionality of the code is to select the columns (variables) of the data from columns 1 to 11 in "White_Wine.xlsx" with which the best classification result can be achieved. In this case, the last column, "Quality" corresponds to the different class of this classification example

in lines 113 and 137 replace "" by "fitcnb". Also, change line 48 to "RandStream.setGlobalStream(s)". Then it seems to work

where to get the whitewine.xlsx

Sorry i want to ask how to fix this error that i get when run this coding tq.
The class RandStream has no Constant property or Static method named 'setDefaultStream'.

thanks sir


Excellent demo. Thanks Sir.

I keep getting en error in 2015 version:

Undefined function or variable 'legacy'.

Error in WhiteWine (line 48)
legacy RandStream.setDefaultStream(s);

xu song




execellent sir



Mr Smart


Excellent presentation and demo on using datasets to do classification.

Ali Ali

Thanks a lot. Fanstastic demo and code.


Updated license

MATLAB Release Compatibility
Created with R2009a
Compatible with any release
Platform Compatibility
Windows macOS Linux

Inspired: Naive Bayes Classifier