Principal Component Analysis / Hebbian-based Max Eigenfilter

Principal Component Analysis and Hebbian-based Maximum Eigenfilter

% TASK 1. Let’s generate 800 random data on a 2-dimensional plane. The data
% are generated as 4 clusters, of which centers are located at (2,2), (-1,-2),
% (2,0) and (0,1). Each cluster has 200 data, of which distances from each
% center are randomly distributed with Gaussian distribution (standard
% deviation = 2, 2, 1, and 1, respectively).

% TASK 1-(a) Mark the generated data with dots (or circles) on a
% 2-dimensional space.
% TASK 1-(b) Conduct Principal Component Analysis based on eigenvector
% analysis. (You may use any library function for the
% eigenvector/eigenvalue calculation.) Show the principal axes and data
% projects on the axes.
% TASK 1-(c) Program and calculate the Hebbian-based maximum eigenfilter,
% and compare with the principal in (b).