Feature Selection Package - Algorithms - Correlation-based Feature Selection
Description
Correlation-based feature selection uses the wrapper method to come up
with a better heuristic for finding the best feature subset.
Usage
Method Signature:
[out] = fsCFS(X,Y)
Output:
out: A struct containing the list of
features that have been deemed useful. The struct contains two fields:
- fList - The list of features that have been selected.
- prf - This means that the smaller the feature weight is, the
more useful it will be to the user.
Input:
X:
The features on current trunk, each column is a feature vector on all instances,
and each row is a part of the instance.
Y: The label of instances, in single column form: 1 2 3 4 5 ...
Code Example
% Using the wine.dat data set, which can be found at
% [fspackage_location]/classifiers/knn/wine.mat
fsCFS(X,Y)
Keyword in Evaluator Framework
cfs
Paper
BibTex entry for:
Feature Selection for Machine Learning: Comparing a Correlation-based Filter Approach to the Wrapper, by Mark Hall and Lloyd Smith of Waikato University
@misc{author = {Mark A. Hall and Lloyd A. Smith},
title = {Feature Selection for Machine Learning: Comparing a Correlation-based Filter Approach to the Wrapper},
publisher = {AAAI},
year = {1999},
URL = {https://www.aaai.org/Library/FLAIRS/1999/flairs99-042.php},
url = {https://www.aaai.org/Library/FLAIRS/1999/flairs99-042.php},
bibsource = {http://www.interaction-design.org/references}
}