PCA and KMC issue

#1
Hello, I'm using these methods for my master thesis. I'm new to this and I'm still learning. Today, I noticed that I get different results from both methods, when the order of the data I use changes. Is that normal? I couldn't find an explanation for this.
 

staassis

Active Member
#2
Why would you think that "KMC" is the proper way to refer to the method? Just because your professor uses this label in his and very his notes? Do try to write in proper English language please. Thank you.

Now, in no way is k-means clustering supposed to deliver the same insights as principal component analysis. Those are very different methods, addressing different goals.
 
#3
Why would you think that "KMC" is the proper way to refer to the method? Just because your professor uses this label in his and very his notes? Do try to write in proper English language please. Thank you.

Now, in no way is k-means clustering supposed to deliver the same insights as principal component analysis. Those are very different methods, addressing different goals.
Firstly, I think it's a proper way to refer to this method because:

a) it's very obvious what KMC stands for
b) that's how it's referred in scientific papers, even in the US

So I'll just continue using that term.

Secondly, I'll write English the way I can, it's not my first language. But you should be less rude and try to behave.

Thirdly, you didn't answer my question, and didn't even understand it, for that matter, so there's no need for you acting so smug about it.

I said that I've tried doing PCA and KMC with some samples, then just rearranged these samples into different order and ran PCA and KMC again, and got different results depending solely on different order of my samples. I can't understand why would rearranging samples in my input file make any difference for KMC and PCA output.

Thanks for nothing.
 
Last edited: