# Cluster algorithm

Hi KNIME team and user,

I am working with knime since few weeks, therefore I am not that strong in KNIME. My main interest is cluster analysis. For my first analysis I used the "Hierarchical Clustering"-node. My question is, what is the algorithm behind this node? How does it works? Is there any literature from where you have implemented the algorithm. It is interesting for me, especially relating to multivariate cluyster analysis.

I hope somebody can give me some advices. Thanks for your efforts!

Regards,

luebke

It's the standard implementation of agglomerative hierarchical clustering. See http://en.wikipedia.org/wiki/Hierarchical_clustering for details.

Thanks for answering that quick. I was not sure for 100%. So am I right, that in multivariate cluster analysis the algorithm calculate like this:

d = sqrt ( (x2-x1)² + (y2-y1)²+ (z2-z1)²+.....)   ?

This depends on the used distance function. In case of the Euclidean distance you are right.

Ah, yes of course, I forgot it to mention.