Kernel Support Vector Machine

If we have an example where the data is not linearly separable, what can be helpful?

• Kernel Reduction

• Kernel SVM

• Kernel Choice

• None of the above

If we have a non linear separable data set, what Weill we need to do for the kernel SVM?

• Map to a higher dimension

• Map to 0 dimension

• Remove all dimensions

• All of the above

One we have our hyperplane in our 3d space example and we have our result we can do which one the following?

• Project it to a 3d space

• Project it to 3d space and remove all points

• Project it directly onto the X axis

• Project it back to the 2D space with our non linear separator

There is a problem with mapping to a higher dimensional space, what is it?

• It can be highly compute-intensive

• It does not have any problems

• The higher dimensional space is not available

• None of above

What is a different approach to help us obtain similar results without having to going to a higher dimensional space?

• The Kernel Methodology

• The Kernel Function

• The Kernel Representation

• The Kernel Trick

What is a different approach to help us obtain similar results without having to going to a higher dimensional space?

• The Kernel Methodology

• The Kernel Function

• The Kernel Representation

• The Kernel Trick