Cluster Analysis Report - Meaning of Avg_Distance, Max_Distance and Separation

[Ricardo380]

Hello again guys!

So, I'm running cluster analysis with K-Centroid Cluster Analysis Tool and I think I'm having pretty good results! But, in order to get a complete detailed analysis (Because i have to present this to a Jury Set and my grade depends on it), I need to know the meaning of the Avg_Distance, Max_Distance and Separation that appear on the reporting snippet - a picture is annexed. I can't find it online, either here or on google...

Edit: Also, what does the principal components 1 and 2 refer to?

Thank you very much in advance, community!

Ricardo

[BenMoss · Accepted solution]

Avg Distance: In average linkage hierarchical clustering, the distance between two clusters is defined as the average distance between each point in one cluster to every point in the other cluster.

^ from google

Max Distance: the greatest distance between points in a single cluster

^ from my knowledge after a quick google

Seperation: the distance of clusters from each other

^ from my knowledge after reading this article: https://shapeofdata.wordpress.com/2013/09/18/modularity-measuring-cluster-separation/

As for your principal components question, I wrote an essay on the subject many moons ago, here's a link, it may be useful and answer the questions you have: https://drive.google.com/file/d/0Bxr8mt2tWVIwWnU2b08wQ3h0aFU/edit

Ben

[Ricardo380]

Hey BenMoss

Thank you for your answer. It was pretty clear and helped me a lot. 

Kind regards,

Ricardo

[RobsonShimada]

I have a question related to your topic, Ricardo380:

What are the distances of the average distance and max distance? I would like to convert this to kilometers, actually, but since the Alteryx Documentation is not informing the units it became pretty difficult to convert.

Thanks in advance,