Distance Matrix Pair Extractor - Clustering

KNIME Analytics Platform
1

Hi community,

I need your help and just can’t get any further. I have been a KNIME user for years, but today I had to give up. :frowning:

I have compared labels with each other and created a distance matrix. I ran the result through the Distance Matrix Pair Extractor (with distance filter). Now I have a list of all similar labels and want to create a cluster.
Here is an example of what the data looks like and what result I would expect:

Data

Col1 Col2
A B
A C
B D
E F
F G
G H
G I

Result
|1|A, B, C, D||
|2|E, F, G, H, I||

I have experimented with the Recursive Loop but can’t get it to work and am getting desperate.

Does anyone have any ideas?

Kind regards and many thanks for any hint!

2

Why not use one of the clustering nodes (like Hierarchical Clustering)?

3

Yes, I tried this and got a slightly different result with the Hierarchical Cluster Assigner and the same threshold value. I can work with it for now or tweak the values.

But I’m annoyed that I can’t figure out how to find these clusters based on the pairs. I have been working with KNIME every day for more than 10 years and it freaks me now that I can’t figure it out. :sweat_smile: