We have the following training data in a 2D feature space for three classes. For class w1, the training samples are { [2

[answerhappygod]

We Have The Following Training Data In A 2d Feature Space For Three Classes For Class W1 The Training Samples Are 2 1 (183.91 KiB) Viewed 67 times

We have the following training data in a 2D feature space for three classes. For class w1, the training samples are { [2; 1), (2;2), (2;4), (3; 2), (3; 3), (4; 1], [6; 1] }, For class w2, the training samples are {(1; 5), (2;5), (3; 5), (3;6), (4; 5] }, For class w3, the training samples are { [6; 5), (7;4), [7;5), [8; 5), [8; 6], [9; 6] }, . where the notation is as follows: [a; b] = [] Your task is to classify a test sample x = [X; 4). (a) Draw the scatter plot. (b) Classify x using K-nearest neighbor classification, where K=3 and Euclidean distance is used as the distance measure. (c) Classify x using K-nearest neighbor classification, where K=3 and Manhattan distance is used as the distance measure. (d) Write the formula to scale the features to [0,1] range. (Do not simply write the formula, specify the numbers/constants in your formula.) Assume that p(xwi) are Gaussian distributions. Suppose that the means and covariances matrices are: m1=[3; 5) and C1=[2 0; 0 1] m2=[4; 2) and C2=[2 0; 0 1] m3=[8; 5) and C3=[1 0; 0 2] (e) Given the Gaussian parameters above, classify the test point x using the Bayesian classifier if the prior probabilities are equal.