The Data Mining Forum 
IMPORTANT: This is the old Data Mining forum.
I keep it online so that you can read the old messages.
Please post your new messages in the new forum: https://forum2.philippe-fournier-viger.com/index.php
hierarchical clustering using Complete Linkage
Posted by:
Nail
Date: August 01, 2019 01:42PM
In hierarchical clustering using Complete Linkage I have an example I did not get how it is done could you please explain how it is done using the example in the below Image
As in the single linkage the same has been done first select 7,10 then 28 35 the last one 20 is different but in both cases the minimum distance got selected
As in the single linkage the same has been done first select 7,10 then 28 35 the last one 20 is different but in both cases the minimum distance got selected
Re: hierarchical clustering using Complete Linkage
Posted by:
webmasterphilfv
Date: August 02, 2019 03:59AM
Hi,
Let's say that you compare two clusters containing several points. You compute the distance between all pairs of points from the two clusters. Then you take the maximum distance among all these distances that you have calculated.
You repeat this for all other pairs of clusters, and choose to merge the two clusters that have the smallest maximum distance.
I think that this is the main idea.
Then you repeat this again to merge clusters recursively.
Best regards
Let's say that you compare two clusters containing several points. You compute the distance between all pairs of points from the two clusters. Then you take the maximum distance among all these distances that you have calculated.
You repeat this for all other pairs of clusters, and choose to merge the two clusters that have the smallest maximum distance.
I think that this is the main idea.
Then you repeat this again to merge clusters recursively.
Best regards