SVM Support Vector Machines
<p>Feb 06, 2011 01:28 PM
 
excellent tutorial!
 
key is the   KERNEL MAPPING FUNCTION.
(default and recommended is the <b>Radial Basis Function RBF</b>)
 
SVM  constructs n-dimension hyperplane that optimally sepates the data into two categories.
 
SVM  using a sigmoid kernel is equiv to a 2-layer perceptron neural network!
 
a predictor variable is called an ATTRIBUTE
 
a  transformed attribute, used to define the hyperplane is called a FEATURE.
 
 
 
 
 
 </p>