a) The data points of a binary classification problem with two features and the class labels...

Transcribed Image Text:

a) The data points of a binary classification problem with two features and the class labels are given below: 20% (0, 2, +), (1,0,-), (-3, 2, +), (0, - 4. -), (1, 5, +), (-1, -6,-). (5, 1,-), (0, 4, +), (-2, 0, +), -} (0.4 s for positio-85546 9. (5,18 (+.) Derive the equation of the decision boundaries for positive and negative examples. Estimate the width of the decision boundary. [2+2+2=6M] b) Consider a polynomial kernel K(U, V) that transforms a two-dimensional feature space U and V into higher order feature space using the function: 540-2 7 and 9/01, [4+2=6M] K(U.V) = (U. V + 1) a) Construct the function o function such 4 85543 14- Q(U). Q(V) b) Given the feature vectors X1(1, 1), and X2 (0, 1), compute their corresponding feature vectors in the new feature space by applying the function (X) obtained above. (1.1) such that x(U.V) = 023/10/0 a) The data points of a binary classification problem with two features and the class labels are given below: 20% (0, 2, +), (1,0,-), (-3, 2, +), (0, - 4. -), (1, 5, +), (-1, -6,-). (5, 1,-), (0, 4, +), (-2, 0, +), -} (0.4 s for positio-85546 9. (5,18 (+.) Derive the equation of the decision boundaries for positive and negative examples. Estimate the width of the decision boundary. [2+2+2=6M] b) Consider a polynomial kernel K(U, V) that transforms a two-dimensional feature space U and V into higher order feature space using the function: 540-2 7 and 9/01, [4+2=6M] K(U.V) = (U. V + 1) a) Construct the function o function such 4 85543 14- Q(U). Q(V) b) Given the feature vectors X1(1, 1), and X2 (0, 1), compute their corresponding feature vectors in the new feature space by applying the function (X) obtained above. (1.1) such that x(U.V) = 023/10/0

Answer rating: 100% (QA)

a To derive the equation of the decision boundaries for positive and negative examples we need to fi... View the full answer