Hey all,
So I am very familiar with the percpetron, and I just recently learned about LDA. I get that it is an algorithm that finds the best hyperplane unto which to project your data for classification... however, I also understand that the perceptron does the same thing.
So... why would someone elect to use LDA over the Perceptron? What advantages or disadvantages are there? I am trying to understand how all this fits together. Is it just another technique to so supervised learning?
Thanks.
[link][10 comments]