Tell me more ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.
up vote 3 down vote favorite
1

Hello I am looking for help on understanding the maths of Fuzzy C Means as explained here: Fuzzy C Means I was hoping for a broken down explantion of the actual math. I have tryed googling for tutorials but it has came up empty. I understand clustering and fuzzy c means and I know how to implement it but I still lack the understanding of the math.

share|improve this question

1 Answer

up vote 0 down vote

I believe it is a version of an EM algorithm like a usual $k$-means algorithm; only that in this case each $x_i$ is allowed to belong to many clusters at the same time. If my intuition is right, you essentially try to maximize the log-likelihood function $\ell(c) = \log P(x|c)$. The EM proof should carry over to show that you can instead optimize $\ell(u, c) = \log P(x|u, c)$ in $u$, then in $c$, and repeat.

share|improve this answer
Hi thanks for the answer but this wasnt really what I was looking for, my maths is alot more basic in what I am looking to understand. For instance a question I asked on cross validated shows a good answer and is really what I am looking for here link – Garrith Graham Sep 13 '12 at 14:57
1  
I thought you were looking for a way to derive the algorithm since you said you understand clustering and fuzzy C means. I will try to give you a more detailed answer when I have time then. In the meantime, if you know about maximum likelihood estimation, you should study EM algorithm for unsupervised learning. (One popular use of EM is k-means algorithm.) – Tunococ Sep 13 '12 at 20:25

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.