Rasa conversational AI on static data

Hi I’m trying to make a Rasa conversational bot which can access a static data-set to give answers to the user. for example, If a user asks “what’s the Revenue generated from abc client”. So the bot will access the data set and output the answer. So how can RASA access a static data?

@Akshit You can build custom actions which can include connections to the database and retrieving data based on what the user asked. For example, your NLU model should be able to extract revenue and abc as entities and set them as slots. Then, in a custom action, you could retrieve those slot values and use them to extract the revenue data for that exact abc client, aggregate the data and send the results back to the user as a response. You can read more about custom actions here.

Hi!!! Thanks for the answer! It helped a lot. So one error i’m getting when running train_online.py i’m getting this error that “from rasa_core import online : importError- cannot import name “online”. and also consider that i want to run multiple actions from a users input in same endpoint server. for example : if user says” Whats the revenue this week and how many clients are added today? So the api client i’m using can give me both the answers! So should i add both these actions in action.py and add the action_revenue and action_client_added in templates and run my bot?will it call both functions at the same time?