Understanding the basics of NLU Engine DialogFlow

Posted By :Harsh Soni |31st October 2019

NLU or the Natural Language understanding engine is gaining popularity with its wide area of possibilities to develop user-friendly & problem-solving applications. DialogFlow is also one of the NLU, an AI-powered tool by google to develop intelligent chat-bots. The main benefit of using it is it provides an easy to understand interface to work with. 

Let's get started by understanding the basic terminologies of the DialogFlow

Agent : 

Agents are the base of the Dialogflow. It's basically a helper for conversation between the user and the NLU Engine. Its conversation flow can be understood in these steps. You can create agents for your custom application or use the pre-build agents to simplify your job.

A user interacts with the Agent we created and sends a request
The Processing Engine(NLU) process the input, and
The response is sent back to the user


Intents are the initiative dialogues turn for the conversation which is created by the bot developers as a way of intimating the Dialogflow service. Let's say the user says this to initiate a chat

"Hey, What's up"
"How are you"

Here are the phrases that seem to be similar and they can be handled by creating a single intent "Greeting" and add all these phrases which a user can use to gree in the "Training phrases" section. In a simple way, we can say that in the Intents we read the user intention and respond back according to the input.

Context : 

Imagine you have passed by your friend and suddenly he says “yes” and you stand clueless why he said yes. a similar thing applies with DialogFlow Chat-bot they also need context to figure out what a particular word or sentence means. Without contexts, the word or sentence could have absolutely no meaning or meant something totally different from the user’s original intent.

Thank you for giving your time to this blog. I hope it was of some help to you.

About Author

Harsh Soni

Harsh is an experienced Developer with multiple tech stack such as PHP, NodeJs, Javascript, Angular. He loves learning new technologies and experimenting on them.

Request For Proposal

[contact-form-7 404 "Not Found"]

Ready to innovate ? Let's get in touch

Chat With Us