Questions and Comments

This slecture to be reviewed by David Runyan

• [DR] (Summary) In the first section, this slecture provides a brief "to the point" presentation of what a discriminant function is, as well as a couple of tricks used in Linear Discriminant Analysis. The slecture then goes on to describe Fisher's Linear Discriminant and how it is used to classify two-class data.

• [DR] The mathematics of Fisher's linear discriminant we presented well enough. In order to improve this slecture, one thing you may want to include is a more intuitive explanation of why Fisher's linear discriminant is set up the way it is (as opposed to any other possible method of choosing c). You could also include a discussion of the shortcomings of Fisher's linear discriminant.

• [DR] One of the equations in the "Set up and Tricks" section is not formatted properly.