Rasa NLU allows you to turn natural language into structured data, but this might not be enough if you want to build a bot that handles what has been said in context and adjusts the flow of the conversation accordingly. Rasa’s open-source solution to handle contextual dialogue is Rasa Core, but there are other tools out there such as Dialogflow (not open-sourced).
Rasa Core uses machine learning to predict the evolution of a conversation,
and does away with the need for tedious and poorly-scaling
It also allows you to implement custom actions in response to the
user message, such as saying something back, modifying a database, calling an
API or handing over to a human. It is by design the natural companion of
Rasa NLU if you want to build conversational bots.