This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Problem Set 4 MAS 622J/1.126J: Pattern Recognition and Analysis Due Wednesday, 25 October 2006 [Note: As usual, please use Python or Matlab when asked to plot data or write a program.] Problem 1: ML Estimation After Dimensional- ity Reduction Download this problem sets data set files from the course webpage. All these data sets consist of 3-dimensional data. There are data sets for two classes, class and class 1. For each class, there are two training data sets, A and B, and one testing data set. a. Use Matlab or Python to reduce the dimensionality of the A training data set for both classes from 3-dimensional to 1-dimensional using Prin- cipal Component Analysis (PCA). As usual, include your program in your answer. b. Use Matlab or Python to reduce the dimensionality of the A training data set for both classes from 3-dimensional to 1-dimensional using Fisher Linear Discriminant (FLD). As usual, include your program in your an- swer. c. Use Matlab or Python to compute the maximum-likelihood mean and variance of the dimension-reduced A training data set for both classes....
View Full Document
- Fall '00