PCA

[anaritam]

Hi
after using
dataSet = prtDataSetClass(features,labels);

dataSet = setClassNames(dataSet,[{'Rest'} {'Index Extension'} {'Fourth Finger Extension'}]);
dataSet = setFeatureNames(dataSet,[{'IEMG'} {'MAV'} {'SSI'} {'VAR'} {'RMS'} {'WAMP'} {'MDF'} {'MNF'}]);

pca = prtPreProcPca;
pca = pca.train(dataSet);
dataSetNew = pca.run(dataSet);

How can I know, which features from the dataSet did the algorithm chose?

As you can see, I have a dataSet with 3 classes and 8 features

Thanks

[kennethmorton]

PCA does not select variables but instead combines them linearly. The weights that are learned by PCA are stored in the property "pcaVectors"

pca.pcaVectors will be and dataSet.nFeatures by pca.nComponents matrix. Each column contains the feature weights that are used to for each resulting PCA component dimension.

[anaritam]

So, it uses all the features and "weights" them in the best possible way?

[kennethmorton]

PCA determines the weights that will maximize the variance in the projection. http://en.wikipedia.org/wiki/Principal_component_analysis