What is the intuition behind the use of the so-called "kernel trick" with support vector machines? There can be more than one right answer.

A. The "kernel trick" performs dimensionality reduction, mapping very-high dimensional data to a lower-dimensional space, where it is easier to model.

B. It is sometimes possible to assume a Gaussian (multi-dimensional normal_ distribution over the data, and use that instead.

C. It s often easier to use a dot product than it is to use a linear classifier.

D. For data where a linear classifier is not a good match, if we map the data into a higher-dimensional space, a linear classifier can still work well.

E. None of the above.