| |
KMEANS k-means clustering object
|
A=KMEANS(H) returns a k-means clustering object
initialized with hyperparameters H.
Hyperparameters and their defaults:
k=2 -- the number of clusters
a.child=distance -- distance measure to use (a distance object)
max_loops=1000 -- maximum number of iterations of training
Model:
mu -- cluster centers
y -- cluster assignment of each training example
Methods:
training: cluster a dataset
testing: assign points to clusters according to the nearest
cluster center
distortion: get value of distortion
Example:
d=gen(spiral({'m=200','n=0.5','noise=0.35'}));
[r,a] =train(kmeans,d)
I=find(r.X==1);clf;hold on;
plot(d.X(I,1),d.X(I,2),'r.');
I=find(r.X==2);
plot(d.X(I,1),d.X(I,2),'b.');
['compare with spectral clustering ']
| |
Reference : chapter 10 (Richard O. Duda and Peter E. Hart) Unsupervised learning and clustering |
| |
Author : Richard O. Duda , Peter E. Hart |
doc kmeans