Related to medicare locator project

Hello, My name is omkar and i’m working on rasa since 3 months now. I must say it is best platform i have ever seen for building chatbot. Well, my question is i’m going through you project GitHub - RasaHQ/medicare_locator: 🏥Medicare Locator - Open source starter pack for developers to build contextual chatbots and AI assistants in healthcare . And in nlu_data.md we have to mention every contain from user so rasa can identify and respond accordingly. But i saw there are some cities you haven’t mentioned in nlu file but present in api. I wanted to know how this is working? How does rasa identify input and respond with hospital name? Thanks.

Welcome to the Rasa community and thanks for the Rasa love! It’s the machine learning of NLU that helps the bot know how to respond to these things – the idea is that you don’t have to define every possible input a user could say, but that NLU starts to pick up on how sentences are structured, e.g. if I say “i need a doctor in ___” there’s a very high probability that it is a city or a zip code. Hope that helps! :slight_smile: