Hi ​@Vaishali , dq rules in Ataccama ONE usually work on the row basis - rule condition would usually evaluate the column values in specific row of data. You can build aggregated rules that would compare specific row of data with aggregated data in the dataset but you need to have a clear idea how you would aggregate it based on what exactly you’re trying to see in the result.

Could you please clarify what is the final goal of the rule?
Do you need to compare the values only between specific categories or between all of them (basically identifying the maximum)?
Thank you!
Ivan

Thanks ​@ivan.kozlov for your response.

I am not looking for aggregate values. I want to compare some of the categories with each other. The comparison may be checking values less than or greater than each other. Or it can be checking if value of category B is x times value of category A.

Hi ​@Vaishali, unfortunately, I don’t think what you’re trying to do is possible, especially using only rules created in web app UI. For the rule condition to be evaluated, all the input attributes should be within the same row of data and the data you need for your rule will never be in one place unless the catalog item structure would look completely different (values for A, B, C, D that have to be compared would be in the same row). 

Going off of this, are there ways to apply rules on a column level? And is there any kind of AI generated automatic classification option when it comes to column level classifications? Thank you 

Hi ​@marella, apologies for coming back a bit delayed here, not for the moment unfortunately. You can see the full list of ONE AI capabilities here and if you’d like please don’t hesitate to share that as a feature request in our ideas section ‍♀️