(Use Matlab)

4. Suppose that we have a two dimensional feature vector and we use a linear discriminant function g(x) w'x+wo to classify an input vector as belonging to one of two classes We decide that a pattern is in Class 1 if g(x)>, and in Class O otherwise (a) Suppose that the decision boundary passes through the point (-1,4) in the -pe with lope 0.5. Find the corresponding values of w and wo (b) Suppose that l.lan w = 1-1 and wo 4 -4 What is the slope of the decision boundary? What is the distance from the origin to the closest point on the decision boundary? (c) Assuming the values in part (b), plot the vector w and the decision boundary g(x)-0 in the x-x2 feature plane. Label which regions of the plane correspond to Class 1 and Class 0. If you draw your figure by hand, be sure to draw it neatly and label it so that the location and shape of the decision boundary can be determined precisely. 4. Suppose that we have a two dimensional feature vector and we use a linear discriminant function g(x) w'x+wo to classify an input vector as belonging to one of two classes We decide that a pattern is in Class 1 if g(x)>, and in Class O otherwise (a) Suppose that the decision boundary passes through the point (-1,4) in the -pe with lope 0.5. Find the corresponding values of w and wo (b) Suppose that l.lan w = 1-1 and wo 4 -4 What is the slope of the decision boundary? What is the distance from the origin to the closest point on the decision boundary? (c) Assuming the values in part (b), plot the vector w and the decision boundary g(x)-0 in the x-x2 feature plane. Label which regions of the plane correspond to Class 1 and Class 0. If you draw your figure by hand, be sure to draw it neatly and label it so that the location and shape of the decision boundary can be determined precisely