I am trying to develop a chatbot using rasa. But I am not getting how rasa using lookup_tables for entity extraction. I had already go through (http://blog.rasa.com/improving-entity-extraction/) link but not getting it. How should lookup_table used for to extract the entity? Also Having confusion about fetching 2 values at a same time from Chatbot. like if user asks to show weather of washington and new york in a single question then how should chatbot understand that it has to fetch weather of both cities at a same time.
Could you clarify what’s unclear about the lookup table?
I think your problem is more related to Rasa Core, your issue is to do with slot fitting if I’m not mistaken and is also an issue I have noticed in the last few months. Your bot needs to be able to undestand when there is more than one city (two slots of type city) and return the weather for both of them.
I am also facing the same issue. Can DIETClassifier be used to get entities from lookup tables?