James T. Metz, Ph.D.
Dear Sir,
MDL QSAR has a component of its non-parametric discriminant analysis
feature that uses a kernel-based algorithm.
The following is from the MDL QSAR help file:
Nonparametric Methods
When the distribution within each group is not assumed to follow a
particular law or is assumed to be other than the multivariate normal
one, nonparametric methods can be used to derive classification
criteria. Among them are the kernel and nearest-neighbor methods. The
former uses uniform, normal, Epanechnikov, biweight, or triweight
kernels to estimate the group-specific density at each observation. The
within-group covariance matrices or pooled covariance matrix can be used
to scale the data.
I have attached a jpeg of the discriminant analysis preferences window
for your interest.
In the interest of full disclosure, I have a consulting arrangement with
MDL, the provider of this software.
I hope this information is helpful,
L. Mark Hall
james.metz~~abbott.com wrote:
>
> QSAR Society,
>
> Does anyone have a strong recommendations for free (or not free)
> programs which process and analyze
> cheminformatic data using support vector machines or so-called
> "kernel-based" learning algorithms?
>
> Regards,
> Jim Metz
>
>
> James T. Metz, Ph.D.
> Research Investigator Chemist
>
> GPRD R46Y AP10-2
> Abbott Laboratories
> 100 Abbott Park Road
> Abbott Park, IL 60064-6100
> U.S.A.
>
> Office (847) 936 - 0441
> FAX (847) 935 - 0548
>
> james.metz###abbott.com
>
> This communication may contain information that is legally privileged,
> confidential, or exempt from disclosure. If you are not the intended
> recipient, please note that any dissemination, distribution, use, or
> copying of this communication is strictly prohibited. Anyone who
> receives this message in error should notify the sender immediately by
> telephone or return email and delete it from his or her computer.
This archive was generated by hypermail 2.2.0 : 2005-11-24 - 10:21 GMT