Nov 2004, at 10:18 PM, [EMAIL PROTECTED] wrote:
From: Berton Gunter [EMAIL PROTECTED]
Date: 5 November 2004 5:08:38 AM
To: 'Dan Bolser' [EMAIL PROTECTED], 'R-help'
[EMAIL PROTECTED]
Cc: Subject: RE: [R] highly biased PCA data?
Dan:
1) There is no guarantee that PCA will show separate groups
Dan:
1) There is no guarantee that PCA will show separate groups, of course, as
that is not its purpose, although it is frequently a side effect.
2) If you were to use a classification method of some sort (discriminant
analysis, neural nets, SVM's, model=based classification, ...), my
On Thu, 4 Nov 2004, Berton Gunter wrote:
Dan:
1) There is no guarantee that PCA will show separate groups, of course, as
that is not its purpose, although it is frequently a side effect.
2) If you were to use a classification method of some sort (discriminant
analysis, neural nets, SVM's,
Dan Bolser dmb at mrc-dunn.cam.ac.uk writes:
:
: Hello, supposing that I have two or three clear categories for my data,
: lets say pet preferece across fish, cat, dog. Lets say most people rate
: their preference as being mostly one of the categories.
:
: I want to do pca on the data to see
I am no expert on this sort of matters, but that has never stopped me from
tossing in my $0.02...
As Gabor and Bert hinted, this is what I would try:
Run randomForest on the data, using sampsize=c(10, 10, 10) and
importance=TRUE, for example. Then take the few most important variables
with