In the previous tutorial, we created a new bot and added a greeting. Now it's time to add some more content. We will start with some basic bot responses to users' frequently asked questions.
In this lesson, we are going to add intents to the Choo Choo bot. We will learn more about the NLP engine and how to update the NLP in your bot, and also how to link intents and messages.
Before we create some more dialogs, we'd like to tell you about the NLP engine first. You see, the Natural Language Processing (NLP) engine is the underlying algorithm that allows the bot to understand what the user is saying. And as each language has its own words and grammar, we have a separate NLP engine for each language!
Understanding language isn't easy: it takes us humans about 6 years and hundreds of examples to understand the most common 20,000 words. It's not so different for computers either. To train an NLP engine, we need huge amounts of data. Luckily, we rely on pre-trained models that have a lot of smarts built in already.
An intent is a specific question from your user or an action they can do. Users will type their question in the bot, which can be recognised by the NLP engine and linked to an intent. For example: an intent can be a question, a statement, an answer to a question, or a greeting. Each intent can be expressed in many different ways, we call these different ways of saying the same thing expressions.
In the example above, the user intent is "How do I sign up for a free trial?". This is then recognised by the NLP engine, which triggers the correct response.
Here are some more examples of intents and expressions:
Intent: book train ticket Expressions:
I want to book a train ticket
I need to go from Antwerp to Brussels
Can I order a ticket here?
Intent: who are you? Expressions:
What is your name?
What can you do?
What should I call you?
Intent: yes Expressions:
Intent: I want to speak to a human Expressions:
Can i speak to a real person?
I want to talk to a human
For this tutorial, we want to give Choo Choo the ability to answer basic questions about itself. To get started, we will create an intent for the question:
Who are you?
On the left side of the screen in the navigation menu, click on
NLP to navigate to the NLP module. Click the
Add Intent and name it
who are you
Now you see that the intent is successfully created, without any expressions added to it (that is what the '0' means below the language)
Now we have to make sure the NLP recognises this Intent. We do this by adding Expressions. Expressions are different ways your users will express one Intent. Just like in real life, there are more ways how you can say a certain thing or ask a question. This is the same in botbuilding.
The more Expressions you add to an Intent, the more accurately it will be recognised. It is crucial for an Intent to have a wide variety of expressions to give accurate results. The more expression you can think of, the better the result of the NLP will be and the 'smarter' your bot will appear.
who are you intent in the Intents pane on the left hand side
The Expressions pane will open in the right. Click on
Who are you? in the open text field
Your screen should look like this:
Add some more expressions by clicking
What is your name?
Can I know your name?
Tell me more about yourself
Please, I'd like to know who I am talking to
How should I call you?
Who is Choo Choo?
Tell me what your name is
Who are ya?
What do people call you?
Are you a train?
Do you have a name?
This will result in the following:
Again, the more expressions you have, the more accurate your bot will be able to respond. Later on, we will see how we can make sure that our bot gets smarter over time, by looking at actual user input once the bot is made public.
Let's try adding another intent and expressions:
Add another intent, like
Greeting and add some expressions:
We have defined two intents now: who are you & greeting.
However, if we were now to say 'Good morning' to the bot emulator, nothing will happen. That is because the NLP is not trained yet, and the intent is not yet linked to a bot dialog. We will work on that in the next steps.
To update the bot, we now need to re-train the NLP. Updating the NLP means that the newly added intents and expressions will be recognised by the bot so we can use them in a conversation.
Update NLP button in the top right corner of the screen:
Select the language you used to add the expressions. You can view the status of the NLP update for each language by clicking on the Update NLP icon.
Update to start the training. This can take a couple of minutes to one hour depending on the size of your chatbot. The more complex, the longer it'll take.
That was a great first step to use the 'Greeting' and 'How are you' intent. The next step is to link these intents in the bot dialogs.
You have now taught the NLP to understand your intents and expressions, congrats! The only thing left to do is teaching Choo Choo how to respond. This means we are going to choose what the response (or flow) should be for each intent. You can do this by adding a new Bot dialog.
Click on Bot dialogs menu item in the navigation pane
Open the General flow
Click on the grey button on top
+ Bot message
who are you as the name
introduction dialog state as the parent (in the Settings tab)
Link the intent to the bot dialog in the bot dialog NLP tab as follows:
Bot Message tab and add a text message that says:
I am Choo Choo, your personal assistant for booking train tickets
Your screen should look like this:
This will result in the folowing overview in the flow:
The image below means that a certain intent is linked to that bot dialog.
If you now say 'Who are you' in the emulator, you immediately get the response that is typed in the 'Who are you' bot dialog.
We have defined the
introduction bot dialog as the parent dialog state in the
who are you bot dialog. Parent bot dialogs do not limit or define the possible flow of the dialogue, they are a visual tool to structure the conversational flow and keep the overview. They make it easier to create complex conversational flows. Bot dialogs can be reached from any point in the conversation by linking a bot dialog to an intent, although you can restrict them too by using Contexts. This mimics the way humans talk, jumping from one subject to another.
As an exercise, you can now add multiple messages to the
who are you bot message. Open the
who are you bot message again and update the single message to show multiple messages:
I'am Choo Choo.
Your train traveling assistant.
You can book a train ticket or ask my support.
After your booking I'll keep you updated about train details so you don't have to worry about your journey.
This makes your bot more user-friendly and human. This will result in the following:
Time to test your bot! Click on
Test your bot at the bottom right to test your conversational flow. To get a feel of your bot's performance, ask the same question a couple times, including different ways of asking the question that are different to the expression you used to train. If a question is not recognized correctly when it should be, you'll have to go back to the
NLP tab, add the questions as an expression, and retrain the NLP model. You can do this as many times as needed, the model will just keep on improving.
Now, you have a bot with the following configuration:
2 intents ('Greeting' and 'Who are you') and their expressions
A bot message 'Who are you', with the intent 'Who are you' and four text messages in it.
You should now be familiar with:
Adding an intent to a bot dialog
Creating intents and expressions
Training the NLP to use these intents and expressions
Adding multiple text messages in one bot message
Testing your intent and messages in the emulator
If any of these topics are difficult for you, revisit them in the tutorial or search on the page in the top right search bar to learn more about a topic.
The next tutorial is about getting user input. We will ask the user for input that is needed for booking a train ticket.