MathGroup Archive 2011

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Support Vector Machines (SVM)

  • To: mathgroup at smc.vnet.net
  • Subject: [mg117781] Re: Support Vector Machines (SVM)
  • From: "Kevin J. McCann" <Kevin.McCann at umbc.edu>
  • Date: Thu, 31 Mar 2011 04:02:17 -0500 (EST)
  • References: <imus8m$i4m$1@smc.vnet.net>

Luis,

I have looked at them, but prefer the somewhat related, but, in my 
opinion, superior Relevance Vector Machine (RVM) originally developed by 
Tipping. The main advantage for my uses over the SVM is that the output 
is the probability that a point in parameter space is in one category or 
another not just a binary result.

I can send you some references (mostly Tipping's) and I do have some 
code in Mathematica as well.

Kevin

On 3/30/2011 5:13 AM, Luis wrote:
> Dear all,
> Does anyone knows any free Mathematica 8 implementation of Support
> Vector Machines for classification?
>
> All my best,
> Luis
>


  • Prev by Date: Re: Filling Plots to the X-Axis for a Range of X Values
  • Next by Date: Re: Fit Gaussian function to histogram
  • Previous by thread: Re: Support Vector Machines (SVM)
  • Next by thread: LaguerreL[n,a,x] with negative n?