1
$\begingroup$

I have a dataframe with multiple features, where I'm selecting 3 features to cluster on. Ex.

Col1 | Col2 | Col3 
-------------------
A    |  1  |   i
B    |  1  |   j
C    |  2  |   k
D    |  3  |   j

I'm trying to cluster such that If there is a matching value in any two columns, it will add that row to the same cluster, for all matches within the values in a cluster.

So from this table, my intended output for a cluster would be:

A    |  1  |   i
B    |  1  |   j
D    |  3  |   j

Is there a clustering algorithm that I can use to achieve this type of clustering?

$\endgroup$
1
$\begingroup$

That is not clustering since the rows are not being grouped together.

It is a filtering problem where the threshold to keep a row is based on frequency in each column.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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