Row Filtering

Hi

I am trying to remove some rows from an invoice data set. An example of the data is shown in the table below where I have identified duplicates based on the supplier mane, PO Number and Abs Invoice Amount:

And this is the output I desire:

I am struggling with the rule needed for a rule-based row filter. If I use (Duplicate Classifer LIKE “chosen” AND Invoice Amount < 0) then I can remove the row 1. If I then use (Duplicate Classifer LIKE “duplicate” AND Invoice Amount > 0) to remove row 2, it will also remove row 4 which is not what I want.

Also, just to add, I can’t group by PO Number and filter on PO Numbers with a zero sum as some PO Numbers have additional amounts associated with them.

Does anyone have any ideas on how I can solve this?

Thanks

Chris