Feature Selection Package Documentation

Kruskal-Wallis is a test of variance using population variance among groups. This function serves as a wrapper to the MATLAB function kruskalwallis.

Method Signature:

[out] = fsKruskalWallis(X, Y)

Output:
out: A struct containing the following fields

W - The distribution at each data point.
fList - The list of features that are deemed useful.
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 ...

% Using the wine.dat data set, which can be found at
% [fspackage_location]/classifiers/knn/wine.mat
fsKruskalWallis(X,Y)

BibTex entry for:

Asymptotic Conservativeness and Efficiency of Kruskal-Wallis Test for K Dependent Samples by L.J. Wei.

@article { wei81,
   author = {Wei, L. J.},
   title = {Asymptotic Conservativeness and Efficiency of Kruskal-Wallis Test for K Dependent Samples},
   journal = {Journal of the American Statistical Association},
   volume = {76},
   number = {376},
   month = {December},
   pages = {1006--1009},
   year = {1981}
}