Getting same answer from bot for user texts that are not in intents

I am a new to Rasa and creating a project. In just starting i am facing problems. I am sharing my files and bot responses. Config.yml

Configuration for Rasa NLU.

Components

language: en pipeline: supervised_embeddings

Configuration for Rasa Core.

Policies

policies:

  • name: MemoizationPolicy
  • name: KerasPolicy
  • name: MappingPolicy

Domain.yml

intents:

  • greet
  • goodbye
  • bot_challenge : {triggers: utter_iamabot}

actions:

  • utter_greet
  • utter_goodbye
  • utter_iamabot

templates: utter_greet:

  • text: “Hey Buddy! How are you?”

utter_goodbye:

  • text: “Bye Bye Buddy”

utter_iamabot:

  • text: “I am a Rasa bot, Developed by AK.” Nlu.md

intent:greet

  • hey
  • hello
  • hi
  • good morning
  • good evening
  • hey there

intent:goodbye

  • bye
  • goodbye
  • see you around
  • see you later

intent:bot_challenge

  • are you a bot?
  • are you a human?
  • am I talking to a bot?
  • am I talking to a human?

stories.md

happy path

  • greet
    • utter_greet
  • goodbye
    • utter_goodbye

My bot response ----------------------

What i want to resolve:

  1. I am creating a complex project which can answer not as per sequence story path based. when story path completes or same intent is triggering then bot failed to respond.
  2. While triggering action for particular intent, bot reponses well but triggering action even for non intent texts also.

please suggest how to do action triggering for specific intents without out story path.

Hello @ak19

If I understand you correctly, you want to build a more complex bot. This is what rasa is made for. Rasa Core will learn to handle unseen situations better when you provide more stories. You can quickly generate those using rasa interactive, or in rasa x.

Concerning your second question: Rasa will always execute some action upon user input, though this action may be action_listen. You have to provide stories to learn from, as I described above, if you want the responds to be better. If you are certain that a particular intent should always trigger a certain action, then you can use the Mapping Policy, see Policies