Data Mining | Week 5

Data Mining Week 5 Answers

link: Data Mining Week (nptel.ac.in)

1. Support Vector Machine is

(A) Maximum aprori classifier
(B) Maximum Margin Classifier
(C) Minimum apriori classifier
(D) Minimum margin classifier

2. Support vectors in SVM are:

(A) Outliers
(B) Subset of testing data points
(C) Subset of training data points
(D) Random points in the data set

3. In a hard margin support vector machine:

(A) No training instances lie inside the margin
(B) All the training instances lie inside the margin
(C) Only few training instances lie inside the margin
(D) None of the above

4. The language multipliers corresponding to the support vectors have a value:

(A) equal to zero
(B) less than zero
(C) greater than zero
(D) can take on any value

5. The primal Optimization problem solved to obtain the hard margin optimal separating hyperplane is:

(A) Minimize 1/2 W^TW, such that y_i(W^T X_i + b) >= 1 for all i.
(B) Maximize 1/2 W^TW, such that y_i(W^T X_i + b) >= 1 for all i.
(C) Minimize 1/2 W^TW, such that y_i(W^T X_i + b) <= 1 for all i.
(D) Maximize 1/2 W^TW, such that y_i(W^T X_i + b) <= 1 for all i.

6. The dual Optimization problem solved to obtain the hard margin optimal separating hyperplane is:

(A) Maximize 1/2 W^T W, such that y_i(W^T X_i + b) >= 1 -α_i for all i.
(B) Minimize 1/2 W^T W + Σα_i (y_i(W^T X_i + b) -1), such that α_i>=0, for all i.
(C) Minimize 1/2 W^T W + Σα_i , such that y_i(W^T X_i + b) <= 1 for all i.
(D) Maximize 1/2 W^T W + Σα_i , such that y_i(W^T X_i + b) <= 1 for all i.

7. We are designing a SVM, W^TX+b=0. suppose Xj are the support vectors and ajs the corresponding Lagrange multipliers, then which of the following statements are correct

(A) W = Σα_iy_iX_i
(B) Σα_iy_i = 0
(C) Either A or B
(D) Both A and B

8. If the hyperplane W^T X + b = 0 correctly classifies all the training points (Xi,yi), where yi = (+1,-1) then:

(A) | | W-1 | | = 2
(B) X =1
(C) W^T Xi + b >= 0 for all i
(D) y_i( W^T Xi + b) >= 0 for all i

9. The dual optimization problem is SVM design is usually solved using:

(A) GeneticProgramming
(B) Neural Programming
(C) Dynamic Programming
(D) Quadratic Programming

10. Stack variables are used in which of the below:

(A) Soft margin SVM
(B) Hard margin SVM
(C) Both in Soft margin SVM and Hard margin SVM
(D) Neither in Soft margin SVM nor in Hard margin SVM