Question

Given that the decision boundary separating two classes is linear, what can be inferred about the discriminant functions of the two classes?
1) Both discriminant functions have to be necessarily linear.
2)Both discriminant functions can be non-linear.
3)At least one of the discriminant functions is linear.

Answer 1

Answer:

1)both discriminant function have to be necessarily linear

Answer 2

Answer:

Option # 1

Explanation:

Linear Discriminant Analysis is a transformation technique that is deployed to reduce the dimensions of the problem under consideration. By reduction dimension, it not only limits the computational cost but also reduces the probability of error.

LDA maximizes the within class variance relative to the variance between classes. Also in this both the discriminant functions have to be necessarily linear so that LDA can be possible