Suppose you have a 2-dimensional binary classification problem, and that the distribution of the observations for each class is plotted in the following figure, showing the observations for one class in red and for the other in blue. O 2 1 0900 0 00 -1 -1 0 1 2 3 4 5 6 a) (10 points) Would it be possible to obtain a classification accuracy close to 100% with a linear SVM? Why? b) (10 points) Would it be possible to obtain a classification accuracy close to 100% with a Radial Basis Function SVM? Why