Chatbot Crash Course for Absolute Beginners - Full 20 Minute Tutorial

Building a Chatbot with Watson Assistant

In this video, the presenter explains how to build a chatbot using Watson Assistant. The tutorial covers intents, entities, and dialogues.

Setting Up Watson Assistant

• To get started, create an IBM Cloud account and log in.

• From the catalog, select AI services and then Watson Assistant.

• Launch a new instance of Watson Assistant and step into your skill.

Understanding Intents, Entities, and Dialogues

• Intents are what the user is trying to do or ask.

• Entities are people, places, or things that you want to extract from the user's text.

• Dialogues respond to the user based on their intent.

Creating General Intents

• In the content catalog under general intents, add "General Greetings" and "General Endings" to your skill.

• Test these intents by clicking "Try it" and ensure that they are responding correctly.

Associating Dialogue with Intents

• In dialogue management, create a new folder called "General."

• Move your general greetings intent into this folder.

• Add responses for your bot to give when it recognizes this intent.

Creating a Chatbot with Dialogflow

In this section, the speaker discusses how to create a chatbot using Dialogflow. They cover creating intents and entities, as well as defining dialogues.

Creating General Greetings and Endings

• The speaker creates an intent for general greetings and endings.

• They test the intent by saying "hello" and "goodbye".

• The speaker creates a node for general endings that triggers when the intent is detected.

• They test the node by saying "goodbye".

Adding Custom Intents

• The speaker discusses adding custom intents to handle menu requests for an ice cream store.

• They create an intent called "menu" and add example phrases.

• The speaker defines a folder to handle menu requests and adds a node called "menu".

• They set up a trigger based on the "menu" intent to respond with available flavors.

Handling Online Ordering

• The speaker introduces entities to handle online ordering.

• No further information is provided in this transcript about handling online ordering.

Creating an Online Ordering Chatbot

In this section, the speaker discusses how to create an online ordering chatbot using entities, intents, and dialogues.

Creating Entities

• Create an entity called "flavors" for the different ice cream flavors (chocolate, vanilla, strawberry).

• Create an entity called "quantity" for the container sizes (500ml and 1 liter).

• Create a pattern-based entity to detect emails.

Defining Intents

• Create an intent called "online order".

• Provide examples of phrases that trigger this intent.

Building Dialogues

• Use slots to prompt for information within a single node.

• Check if the user has provided all necessary information (flavor, quantity, email).

• Store collected information in context variables.

• Condition responses based on collected information.

Conclusion

The speaker provides a step-by-step guide on creating an online ordering chatbot using entities, intents, and dialogues. The chatbot prompts users for necessary information and stores it in context variables before providing confirmation of their order.

Building a Chatbot End-to-End

In this tutorial, the speaker demonstrates how to build a chatbot end-to-end using intents, entities, and dialogues. The chatbot is built to handle orders and can be integrated with Facebook Messenger.

Extracting User Email

• When triggering the pace of dialogue, only matched values are received.

• To extract user email from input text, use reg X to take it out and return it as user's email back.

• After testing the chatbot by ordering ice cream and providing an email address, the extracted email is displayed.

Deploying Chatbot

• To deploy the chatbot on a site or integrate with Facebook Messenger:

• Hit "Add Integration" and select Facebook Messenger

• Follow steps one through five for full integration

Conclusion

• Through this tutorial, viewers learned how to build a fully featured chatbot that can handle orders using intents, entities, and dialogues.

• Viewers are encouraged to like, share and subscribe if they found this tutorial helpful.