Counting number of unique rows


I have a big dataset consisting of applications (Adobe Reader, Photoshop, etc.) and user informations (name, ID, department). I want to calculate with KNIME how many users per department are using a application, so that I am able to say in the end, department X has 100 users using application Y while department Z has 10 users using application Y. Do you guys have an idea?

Thank you very much

Take a look at the GroupBy node, you can group by department and software, add an count aggregation to get the totals.

For the sake of completion, I was trying to count unique rows for all columns meaning that I am putting all my initial columns for grouping, but then I had no column left to do the count.

My solution was to use the constant value column node, to add a column of 1, such that you can use this column for the count aggregation.