If you’ve used Siri, Alexa, or the Google Assistant, you’re familiar with how chatbots work. It’s a piece of software that allows you to interact with your computer by conversing with it. At least, it try to have a chat with you. Historically, chatbots were quite complex and not all useful.
Now things have changed dramatically. We use our voice assistants all the time, and they almost always get what we mean. Log into almost any commercial website and pop-up messaging app that may not have a real person behind it, offer to help.
Chatbots are revolutionizing the way we use computers, but modern chatbots are complex to code and labor-intensive to train. That’s why we now have so many companies offering chatbot builder to make the process (relatively) easy.
The AI technology leader that runs the world’s best chatbots, IBM, actually offers a chatbot builder that you can try for free. It is called IBM Watson Assistant and allows anyone to take advantage of one of the leading AI systems in existence.
You may have heard of Watson before, specifically when it beat human opponents in games Risk! It is more than just a toy, you can build a workable chatbot using their technology and then integrate it with your own personal or business use cases.
The free account allows up to 10,000 API calls to Watson Assistant per month. For most personal users, this is more than enough. It’s definitely enough for what we’re going to do here!
Break it down to the basics
Before diving into building a chatbot with Watson Assistant, let’s talk about what we want to achieve.
The basic purpose of a chatbot is to explain what the user is saying and then figure out what to do with it. In Assistant Watson, there is something called intent is basically the action the user wants to achieve.
For example, someone’s purpose might be to find out if you have a particular item in stock. In that case, Watson Assistant can be integrated with your database and then pull the correct inventory levels from there.
Intentions are made based on entitycontains relevant information Watson will use to respond.
When building your chatbot, you go through a three-step cycle:
- Create Intents and Entities
- Build a conversation tree
- Trying out your chatbot!
In this tutorial, we’ll go through one iteration of this loop, to show you the core building block of a Watson chatbot.
Begin
The first thing you want to do is go to Watson’s Assistant page and register. We used the free “Lite” plan that allows you to build any chatbot you want with no strings attached. Press only Get started for free And follow the instructions.
Once you’ve completed all the digital paperwork, you’re done Introducing IBM Watson Assistant Page. Now click Create a workspace.
Now click Create.
Now just Name your workspace and add a description.
Before we add our custom intents, the first thing you should do is add some general intents that IBM prepared for us. This saves you from having to teach your bot the basics of conversation from scratch.
To do this, in Intents, click Content catalog and then click Add to workspace next to the Shared Category.
The bot is now capable of understanding common dialogs, such as greetings. To see how this works, click Dialog box tab and then click Create dialogs. Two conversation buttons will be created automatically – Welcome and Anything else.
Click Welcome to expand it.
Here you can see the basic structure of the dialog button. If the bot recognizes a certain condition (such as an intent) it will respond in the way you specify.
Although we didn’t create a custom intent in this case, go to the Intents tab and see the intent named #General_Greetings. Just click #General_Greetings in the list of intents to open it.
This is exactly how you implement your own intent. Give it a name, description, and more ways you can think of for users to indicate this intent. Here IBM has done the work for us, but it’s completely clear how it works. You can also modify this existing intent by adding more examples. Perhaps in your own local dialect.
We’ll be using this intent as part of our chatbot dialog, so back to dialog tab and click Create dialogs. Our custom dialog will happen between the Welcome and Anything else buttons. All you have to do is click Add button.
We will call this button Friendly greetings and will use #General_Greetings intention to power it. So under If the bot recognizes just enter the name of the intent and select it from the drop down menu.
Now we leave our response as “text” and write what we want the bot to say in response to our greeting.
You can close the dialog button by clicking X button. We should now have a bot that can respond to a basic greeting. Click Try it at the top right of the page to check out the bot we’ve built so far. Type Hi in the chat bar that pops up and see what happens.
OH! Watson recognizes our intent to greet it and then uses the rules we specified in the dialog builder to respond. Congratulations, you’ve just gone through a full cycle of building a chatbot!
Advanced Features
While you can build a pretty awesome chatbot using this loop of basic steps, of course, Watson Assistant is much more than that. Using all the features of this tool, you can build a bot that handles bookings or provides users with a list of products that are in stock.
The sky is the limit and you don’t have to be a high-tech genius to go. IBM has tons of videos and tutorials to make you a chatbot master. Now that you’ve got a taste, nothing can stop you!
