I am new to machine learning and I am trying to do classification for some sets of data. I need to use PCA.
My questions are: Does PCA only transfer the data to the new coordinate system?? And Do I have to use another method such as Clustering or SVM to do the classification after PCA?? If not, I really appreciate to guide me how to do the classification with PCA.
Thanks
Yes, PCA is just a preprocessing technique, which rotates/scales your coordinate system (and can drop some dimensions). Consequently this is not clustering nor classification tool. In order to use it with any other technique simply use it one after another - PCA is simply a data preparation step, thus transform your data before applying any clustering/classification. Remember to store your pca object so in the future, during test time, you can preprocess test points with the same pca.