Suppose we are given two classes class-1 and class-2 and the mean of the two classes are $μ_1$ and $μ_2$ respectively before projection. Both of their variance-covariance matrix is also provided before projection. After maximizing $J(W) = \frac{\det(W^T S_b W)}{\det(W^T S_w W)}$ we are given two eigenvectors. We now need to determine, after transforming test data with eigenvectors, to which of the two classes they belong too. How to approach this problem? Any suggestions would be helpful.
The link of the full problem is given Here: Problem
