Java Applet for Laplacian SVMs 

Click  on  "Load toy1".   The  large  colored points  are labeled examples  and the small white points are unlabeled examples. Press "Run".   You will see the decision surface obtained by an SVM using just the labeled examples. Now change the options in the text area below to "-t 2 -c 100 -l 1000" and  press "Run".

This is the decision boundary obtained using Laplacian SVM. You can upload  your 2d dataset or interactively construct a dataset by using the "class:1" button.

The toy examples demonstrate the "cluster" and the "manifold"
assumptions for semi-supervised learning. This applet has been adapted from the Libsvm package.

 

                 

 

Options:    

-t kernel_type : set type of kernel function (default 2)
    0 -- linear: u'*v
    1 -- polynomial: (gamma*u'*v + coef0)^degree
    2 -- radial basis: exp(-gamma*|u-v|^2)
    3 -- sigmoid: tanh(gamma*u'*v + coef0)
-d degree : set degree in polynomial kernel 
(default 1)
-g gamma : set gamma in radial basis kernel 
(default 1/k) 
-r coef0 : set coef0 in polynomial/sigmoid kernel function  
-c cost : set the parameter C of SVM (1/2*gamma_A)


For Semi-supervised learning, include the following parameters, 

-l deformation parameter : ratio of intrinsic and extrinsic       regularization (default 0: standard supervised SVM) 

-k number of nearest neighbors used to contruct the Graph laplacian (default : 6)