I want to understand the kernel selection rationale in SVM.
Some basic things that i understand is if data is linear then we must go for linear kernel and if it is non-linear then others.
But question is how to understand that the given data is linear or not(specially when it have many features).
I know by cross validation i can try and feed different kernels and see the output, whoever performs best to be selected.
But anyway to have some early indications ?