Association rules criterion

rpdmnr

Hello. 

I've been working on a project, and I have to analyse the association rules of a database containing texts. The process works fine, but I was wondering which criterion is the right one to use for text data, confidence or lift?

The association rules of both algorithms make sense; the ones from the lift criterion are more basic and not so impressive. Why is that? Because with the lift criterion we consider the most correlated attributes?

Thanks in advance

sara20

@rpdmnr

Hello

Take look at this link 
https://academy.rapidminer.com/learn/article/cross-selling-do-you-want-fries-with-that


Best
Sara