Updated: Dec 16, 2022
Rasa is an open source Machine learning framework for automated text and voice-based conversations i.e, you can easily access and use it in your project anywhere you want.
Rasa is helpful in understanding messages, holding conversations and connecting to messaging channels and APIs.
Suppose I have created a chatbot and that chatbot has made so many conversations with so many different users so in that case I want to keep all the records of conversation that my chatbot has done the conversation with any other person.
So rasa is capable of holding all the conversation that stored in the database and also it can connect to the messaging channel so whatever messaging channel you are having with you, you can connect that to rasa so that in future if you want to connect to any of the messaging channel for the conversation part so you can use it there with your own rasa chatbot and also with the API.
Why to use Rasa:
● All your training data is not dependent on Google, Microsoft, Amazon, or Facebook.
● Machine Learning models are not one-size fit all, you can customize the model for your training data.
● Rasa runs anywhere you want, so you don’t need to make any extra network requests for every message.
Components of Rasa:
Rasa NLU: It is an open source natural language processing tool for intent classification and entity extraction in chatbot. It is also called the ear of your bot.
For example: if you want to book a cab for 5 people,
Intent - cab booking
Entity - information about your context
Rasa Core: it is the brain of your chatbot. it places a major and most essential part in generating the reply messages for chatbot. It considers the output of Rasa NLU (intent and entities) as an input and applies machine learning models to respond with a bot reply.
A tool which is specifically designed to make it easier for you to share your chatbot to the user as soon as possible and collect the conversation between the users and your bot and use the collected conversational data to improve your bot.
To build Rasa project:
To build your first rasa project first you have to install rasa by following command in your Anaconda prompt.
$ pip install rasa
After installing rasa you will also need to install spaCy English language Model.
$ pip install spacy
$ python -m spacy download en_core_web_sm
To train your model you just have to run the following code in your Anaconda prompt. The model will train with a default data come with the rasa package.
$ rasa init
Author: Abhishek Singh