2. (9 points) Penalty points obtained in two DISCO committees of various students at IITD and their hostel expulsion status active/ suspended (A/S) is presented in Table 1. The dataset is linearly separable. As it is desired to train a SVM to classify these data, the following need to be done: (a) Suitably formulate dual optimization problem for SVM using the data and clearly present the following in matrix form: (i) objective function; (ii) equality constraints, and (iii) inequality constraints. You have to numerically compute show the each term. Write the steps will not give any credits. No need to solve the optimization problem. (b) If the solution of the above optimization problem (a) yields the decision vari- able vector as [2 2 0 0 0 0 4 0] (same as in the sequence of data), compute the equation for the separating hyper plane (c) If a new student has got 1.75 points in Disco 1 and 2.5 Disco in Violation 2, will his/her hostel status will remain active or suspended? A/S Table 1: SVM data Student Penalty points S.No Disco 1 Disco 2 (i) 2 3 (ii) 3 2 (iii) 4 2 (iv) 5 1 (v) 1 0 (vi) 1 2 (vii) 2 2 (viii) 1 1 S S S S A A A A PN