What model is being used in RASA to identify role and group in rasa entity extraction

Rasa Open Source
1

I was going through Entities Roles and Groups i could see the the output identifying roles like below “text”: “Book a flight from Berlin to SF”, “intent”: “book_flight”, “entities”: [ { “start”: 19, “end”: 25, “value”: “Berlin”, “entity”: “city”, “role”: “departure”, “extractor”: “DIETClassifier”, }, { “start”: 29, “end”: 31, “value”: “San Francisco”, “entity”: “city”, “role”: “destination”, “extractor”: “DIETClassifier”, } ] }

what kind of model is being used to identify role , since the same word has two tags city and destination. are they training two entity models to get this output

2

hi @AviPaul! Essentially it’s the same model (a CRF) which does the entity extraction in the first place, it’s just a second CRF which is integrated into the DIET architecture. You can also learn more from Tanja’s post here

3

thank you @amn41 will go through the link