terms intersection

Regardless of using the reference row filter or the set operator to find terms common to training and test documents, there's a particular term that is found in training but not testing.

What could be the reason? As of now I have to remove this term from training to build classification models. Is it a case of duplicate rows in training ?

What term is it? Can you provide data or a workflow? When comparing terms, the tags are also taken into account.

Cheers, Kilian

thanks killian i fixed it somehow but can’t remeber what i did a while back. will share if i remember

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.