Given a general problem xi$($i $=1,2,...$N$)$ of m dimensional features labeled as L

number of classes, you are quested to design an SVM solution for classification

task. You need to draw a figure to show how you design the classifier:

$1.$ How you plan to use the data given? Such training, validation and test set?

$2.$ What are the key parameters in SVM you need to tune?

$3.$ What is your plan to achieve the optimized solution?

$4.$ How are you going to test the robustness of your model? additionally, In the m dimensional features, how you going to select the most $$Powerful$$ features?