# How is the multi-class support vector machine classifier implemented? [closed]

I am using support vector machine (SVM) method with Classify to do multi-class.

Can anyone tell me the reference (paper) used by Wolfram Mathematica to develop the SVM method for the multi-class version?

We can get the version of multi-class method of SVM using Classfier Information

 ClassifierInformation[classifierSVM]

Association[
"Method" -> "Support vector machine", "Number of classes" -> 8,
"Number of features" -> 640, "Number of training examples" -> 1600,
"Number of extracted features" -> 340,
"Kernel type" -> "Radial basis function",
"Soft margin parameter" -> 1,
"Multiclass method" -> "One-versus-one"].


I this case, the Multiclass method is "one vs one"

## closed as off-topic by m_goldberg, dr.blochwave, user9660, Jason B., MarcoBJan 20 '16 at 13:18

This question appears to be off-topic. The users who voted to close gave this specific reason:

• "The question is out of scope for this site. The answer to this question requires either advice from Wolfram support or the services of a professional consultant." – m_goldberg, dr.blochwave, Community, Jason B., MarcoB
If this question can be reworded to fit the rules in the help center, please edit the question.

• Does Mathematica ever disclose the details of the implementations? I don't thinks that information is available other than what can be guessed from Method names, as in "RandomForest, classify using Breiman-Cutler ensembles of decision trees" – rhermans Jan 19 '16 at 15:41
• @rhermans, I don't know. – BetterEnglish Jan 19 '16 at 16:16
• You might want to write to Wolfram Research and ask them. I can recall someone here on mathematica.SE recommending this and saying that they (Wolfram) were very forthcoming with their sources/papers. – Sascha Jan 19 '16 at 18:53