Service Analytics

6.Principal Component Analysis

Eigenvector und Eigenvalue

 
 This means that the desired u is an eigenvector of the covariance matrix C and λ is its corresponding eigenvalue.
u is the vector which captures the largest variance of the original dataset. We call it the first principal component u1 with eigenvalue λ1. The second principal component u2 is found in exactly the same way. And so forth until um is found.

Diskussion