Row Filter based on Multiple Columns (Filter out Rows where all Columns have Missing Value)

[Er3n]

Dear Knime Community,

I hope you can help with the following problem. I have a table where each column represents a KPI and each row a unique ID. I’d like to filter out the ID rows where none of the columns has at lease one value.

In the screenshot below the rows for the IDs 4, 7, 11 and 15 should be filtered out:

The result should look something like that:

Ideally the workflow could generate two tables: one without the filtered out rows and one with only the filtered out rows.

Any ideas on how to approach this?

Thanks a lot in advance!

Best
Eren
Example.xlsx (10.3 KB)

[elsamuel]

The first thing that comes to mind is using a Column Aggregator node to concatenate the row files (minus any missing values):

This is the result:

Then you can use a Row Filter to remove any row that has missing values in the Concatenate column:

[mlauber71]

@Er3n I have once created this solution to apply and combine several rules to a collection of columns and then aggregate the results. Will have to see if I can adapt this to your case later

Help with rule-based row-filter - according to excel formula

@DennisMB welcome to the KNIME forum. I have prepared this example on the KNIME hub. You can write a column expression testing a single column for your conditions (0 = a negative condition is met, 1 = the positive condition is met, 2= something else, maybe also a 1?) [image] Then for the fun of it I have built a system where you could apply the rules to all columns (you want). If we have one negative in a set of columns it is a negative overall and we would need at least one positive mark in …

[Er3n]

Thanks a lot! It works

[Er3n]

Thanks @mlauber71 Will take a look at it