I have a problem statement of classifying emails over a big collection of 100-150 categories. And I want to use LUIS (Language Understanding) Cognitive Service provided by Azure as an API.
But as I tested LUIS over a bunch of categories, say 5-10. I found I'm not getting correct predictions. Maybe the reason is not providing the adequate amount/ type of utterances.
So I would like to know :
Is it possible to do so using LUIS?
If possible, what approach should I follow to get correct predictions?