Skip to content

Genetic algorithm for feature selection

1 message · Rolf Wester

#
Thank you very much for your reply. I found the information on your home 
page very useful.

What I want to do is a PLS regression of a data set with 60 features for 
calibration purposes. 
In order to optimize the performance of the calibration I have to find 
out what features to use
in the PLS regression. Although this is what PLS-regression does but I 
found some articles
(R. Leardi, "Genetic algorithms applied to feature selection in PLS 
regression: how and when
to use them", Chemometrics and Intelligent Laboratory Systems 41 (1998) 
195-207) that claim
that a genetic optimization algorithm applied to feature selection 
could  further improve the calibration.
I found Python code that does exactly this but although I'm much more 
experienced in programming
in Python I think it's probably better to use R for this task. So I'm 
going to implement the genetic
algorithm described in the papers by Leardi in R.

Regards

Rolf
Patrick Burns wrote: