+50, Great job!

Also, In the “To sum up” section, statement 1: ”the higher the C, the more penalty SVM was given when it misclassified, and therefore the less wiggling the decision boundary will be”

don’t you think it should be the opposite?

The higher C(inverse of lambda, multiplies with objective function) denotes low lambda (multiplies with regularization term) and hence less penalty which could lead to the wiggle decision boundary.

The high value of C = less regularization = more prone to overfitting.

Correct me if I am wrong.