LDA vectors coefficient interpretation -


i trying interpret/quantify coefficients of vectors obtained after lda.

let's obtain eigenvector (unitary)/score 2 classes lda, such as:

0.1348 0.2697 0.4045 0.5394 0.6742 

the last dimension important in ability discriminate, right ? can quantify 0.6742^2 /1=0.4545 %, meaning dimension contribute 45% of discrimination "power"/ability ?

or 0.6742/2.0226=0.3333 contributing 33% of discrimination "power"/ability ?

or neither of right , how quantify if possible.

i sorry if trivial, couldn't find clear answer yet on internet.

thank you,

tual


Comments

Popular posts from this blog

android - InAppBilling registering BroadcastReceiver in AndroidManifest -

python Tkinter Capturing keyboard events save as one single string -

sql server - Why does Linq-to-SQL add unnecessary COUNT()? -