How to map raw conversations into stories for training Rasa Core?

Rasa Open Source
1

After reading the docs and some tutorials. I do not understand how to create rasa stories from raw real conversations. For example, given: the following conversation:

User                                                           Agent
Hello        ->                                            Good day! What can I do for you today?
I am looking for a restaurant within 500 meters       ->   I found a restaurant called Zur Letzten Instanz based on your search parameters
What is the address of this place?       ->                The address is Waisenstrave 14-16, Berlin
And what is the raiting       ->                           The raiting is 4.1
Is it currently open?       ->                             The place is currently closed
Thanks       ->                                             Happy I could help!

How should I map it into rasa stories.md format? In this tutorial is not very clear how to do this process.

1 Like
2

Hey @J.Doe. For now, you would either have to convert the conversations (at least some of them) to Rasa stories by hand or use interactive learning generate to generate training stories in a Rasa format by talking to your bot.

1 Like
3

Hi, thanks for the reply! I forgot to mention that I also have the actions, slots, entities, intent associated to each response and query (user/agent)

4

Hi j.Doe ,Did you get any solution for that ,if conversation data is in gbs , what should be approach. is there any solution