Finding the closest axis for each principle component.

http://www.google.co.in/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0CC4QFjAB&url=http%3A%2F%2Fsatoh-lab.ex.nii.ac.jp%2Fusers%2Fledduy%2Fpub%2FLe-Satoh-ICAPR05.pdf&ei=LkgxUpLGJMjirAfPoYCQBQ&usg=AFQjCNEK8X-sk7GByLsTzO-jiHHhl3M0OA&sig2=JFRdb8FlPBcIA7n-1XEPrQ&bvm=bv.52109249,d.bmk

I am trying to implement the feature selection part mentioned in the above paper.

I have computed the principle components and sorted them in order of magnitude of eigen values, as required.

The problem: For each principle component find the axis that is closest to it. I don't clearly understand what the authors mean by this statement. If anyone could explain and help me out with the code it would me immensely helpful.

Answer 1

[Jan]

If you do not understand, what the authors mean, ask them. It would be much less efficient, when the contributors of this forum read the paper at first before they solve your problem. But the authors of the paper are surely professionals in the concerned topic.

It would be possible to assist you with the coding, if you post what you have done so far and ask a specific question concerning the code. But it is very unlikely that anybody will do the research for you and can fix your code without seeing any line of it.

[sidra]

@Jan Simon : Thank you