I would like to know understand more how to create a NLU pipeline and their drawbacks…
One thing I am trying to acheived, is to get 4 or less entities extract based on an answer.
The question could be: How many adults and children would you be ?
The answer could be: Two adults and two children; Two adults and one child; One adult and two child; One adult and one child.
And probably many other variants
How to differenciate adults/adult and child/children and extract their count?