Re: Principal Component Analysis and finding important weights

Hi Erdem,
I do not know exactly the solution for your question.
But if try with Fuzzy logc further , it may help you. I tried the same
for some feature identification of hight impact signals on automobiles.
it worked for me.


Regards,

Sathish Athreya

erdem wrote:
Hi,

After having the PCA model, how can we give weights to the features of
the problem?
(the weights of the features gives the importance of the features of
the example data set)

and is it reasonable to use PCA based weightening for finding the
importance of a regression or optimization problem?

Thanks

Erdem

.

Relevant Pages