[R] 5 binary_class models vs one 5-class model

From: Weiwei Shi <helprhelp_at_gmail.com>
Date: Tue 26 Sep 2006 - 21:09:09 GMT


Hi,

I apologize this question is not very r-related, but believe many people using R are expertised at or interested to know the answer to the following question.

I am having a problem in classification. In bioinformatics study, we always ends with a limited size of samples. While in algorithms, some specific algorithm cannot handle modeling with more than 2 classes problem. For the time being, not considering those limitations, I just have a general question like this:

suppose I have a problem for classification, which involves 5 classes. I am wondering if there is a general research comparison on which approach is more accurate: building 5 binary_class models or building one 5-class model (suppose cost (penalty) is same when accuracy is estimated).

An extended or more practical question, in bioinformatics, if you do not have many samples but you are having such problem, what approach will you take?

thanks,

-- 
Weiwei Shi, Ph.D
Research Scientist
GeneGO, Inc.


"Did you always know?"
"No, I did not. But I believed..."
---Matrix III ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Received on Wed Sep 27 07:22:01 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Tue 26 Sep 2006 - 21:30:06 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.