Things To Know about  RASA : Dialogue Management

Sit back ,Relax , Read, Learn and Enjoy....

Every new learning always begins by Asking the right question …Right ?

Okay fine..Lets rise few Questions !

• What is RASA ? and Why?

  • RASA is a open source ML tool to build Contextual AI assistants & custom AI chatbots using Python , Natural Language Understanding (NLU) and Core.

  • Rasa gives complete freedom to configure NLU, Core, Integration, Deployment, etc.

  • Check Out more on docs of RASA.

• What is Contextual AI assistants?

  • It goes far beyond Simple FAQ interactions.

  • It will consider the context of what has been said before which is very important part of every natural conversation.

  • Evolution of Contextual AI assistant

    • Last 10 Years - Notifications

    • Today - Chatbots - FAQs

    • Trending - Contextual Assistants

    • In Few Years - Personalized assistants

• What is Chatbot ?

Chatbot is software that simulates a conversations between user and computer.

• What is NLU?

  • It is a branch of AI  Which understand input in the form of sentences using text or speech.

  • Its a kind of Ear for our Assistants.

  • It takes user input in an unstructured Human language and extract structured data in the form of Intent and entities.

• What is Dialog Management (so called core) ?

  • It is a computer system that is able to engage in an interactive dialogue with a human user about a particular topic. Metaphorically, all interaction can be seen as a kind of dialogue."

  • It is the brain of our assistants.

  • It makes the decision on How an assistance should respond based on the specific state of the conversation as well as the context.

  • Core learns by observing the patterns from conversational data between the user and the assistants (So called stories).

• How core Learns to keep the context?

  • A ML Model under the hood observes what the conversation is currently about and also looks what has been said before.

  • In addition to this, the model takes into account the details extracted by the NLU model.

  • some entities can have a direct influence on how the dialogue management predicts the next action.

I think The Right questions can be asked Only by doing some actions...right?

So Lets do Some actions...

How to start developing an basic assistance with Rasa?

The quickest way to do it is using the Rasa CLI command rasa init which will automatically creates a Rasa project for us with some example training data and all the important project files needed to build an assistants....

Lets dive.....

In RASA playground..we are going to do that.....