Question 2 (3 Marks) Assume you have two datasets of (x, y): . D1 = {(-1, +1), (0, -1), (1, +1)}. . D2 = {(-1, -1), (0, +1), (1, -1)}. If you use feature q(x) = x, neither dataset can be linearly separated. You need to define a two-dimensional feature p(x) to fix this such that: . A weight vector wi can classify D, perfectly (i.e., w1 . (x) > o if x has label +1 and W1 . Q(x)