In this module, we're going to create a chatbot for the fictitious flower shop.

The virtual assistant we are creating uses Watson Conversation,

which is hosted as a service on Bluemix, a platform as a service provided by IBM.

If you're not familiar with the concept, you can think of it as a playground where

developers and business users can create applications and

run analysis by leveraging the series of services hosted in the cloud.

This includes Watson Conversation and other Watson services as well.

So our first step is to register for a Bluemix account.

I've provided the link under the video which you can click to register.

This step is also part of the lab for this module, so

you don't have to pause the video now to complete the step.

Registration is free, no credit card information is requested at this stage.

We can use the trial for free, and without commitment.

The free trial currently includes up to 2 gigabyte of run time memory, up

to 10 services, and access to any service in the catalog that offers a free tier.

We're in luck, since Watson Conversation offers such a free tier.

When the trial is over, you'll have the option to continue using the services for

free by providing your credit card information.

Now, if they're free, why would you need the credit card?

Your card won't be charged unless you decide to use premium services, or

exceed the free allowances.

To help you out, Bluemix provides a cost calculator to estimate costs beforehand,

as well as a usage dashboard to see current usage.

In practice,

you only end up paying when you're ready to grow in scale services you use.

Registration only takes a few moments.

Just make sure to verify your account by clicking on the link within the email

Bluemix sends you.

If you don't see the email in your inbox, check your spam folder as well.

Once verified, you'll want to login into Bluemix.

Since this is a new account,

we are asked to name our organization, using your company name is a good idea.

In my case, I'll simply take this address and build your own chatbot.

Next, we are asked to create a space.

A space will be a collection of applications and services.

You might, for example, want to use spaces as environments, such as development,

test, staging, and production.

Others might want to use them simply as namespaces, a way to logically divide

applications and services for different departments within the same organization.

In our case, we'll do it live, and simply select production.

In the top right, we can review the current region, organization, and

space, where we'll create apps and services.

By default, applications and services will be deployed in the US.

But if your customers are mostly located elsewhere,

it might make sense to change it to a closer cloud location to improve latency.

From this page, we can access the whole Bluemix catalog.

Typically, you'd want to create an application, like a Python or

Node.js web application, to act as the chat user interface.

And then bind it to a Watson Conversation service, and

optionally, to other services as well, such as databases.

However, since we'll be deploying our chatbot by a WordPress plugin,

in this course we only need the Watson Conversation service back end.

We won't need the front-end application, and therefore,

no ability to program is required to follow this course.

Give it a name that you can easily recognize if you end creating many

services.

For example, Watson Conversation Flower Shop.

We don't need multiple credentials, so we can leave the credential name as it is.

Next, we'll need to select a pricing plan, the lite plan will do for now.

But as I mentioned, if you deploy your chatbot in production,

you might want to consider the pay-as-you-go standard plan.

Now that we have added a Watson Conversation service,

we can launch the tool to start working with Conversation.

The workspace section is shown to us,

here we can look at the car dashboard workspace provided as an example.

As you can see, we have a bunch of intents, entities and

a dialog, all three components of every Watson Conversation chatbot.

By clicking on Watson Conversation in the top left,

we go back to the workspace list.

Here we can click on Create, and generate our own workspace for

the flower shop chatbot.

We are going to call our chatbot Florence, to give it a bit of personality, so

let's call the workspace Florence Chatbot.

For the description, we can specify what the chatbot will do,

including a hint at its corporate responsibilities.

For example, we might enter something related to the flower shop chatbot

that provides assistance on flower suggestions and delivery information.

As mentioned before, we can select a different language for

our chatbot if we prefer.

In our case, we'll use the default, which is American English.

As you can see, they're empty, but we have Intents, Entities, and Dialog tabs.

We can close the Help section and re-open it by clicking Show Help.

There's also a link to a forum if we run into trouble,

and would like help from the community.

Don't forget that there is also a discussion forum in this course,

should you have any questions.

In the top left, we can see that we are in a Watson Conversation service

in the Florence Chatbot workspace, and in the building phase.

Aside from Build, we also have Deploy and

Improve, which will be covered in a module on deployment of chatbots.

We can reduce the sidebar again by clicking on the X.

By clicking on this table icon, we're sent back to the list of workspaces.

To jump right back into our Florence workspace, we simply click on it.

By default, when we click on a workspace, the Intents tab is shown,

since it's usually the first step in creating a chatbot.

In the top right, we have three icons,

the first one covers information to learn more about the service.

The second button logs us out.

And finally, the third icon is the Try it out panel, that I've shown you before.

Obviously our workspace is empty at the moment, so there's nothing to try.

But we'll get more familiar with this panel in the next video of this module.

Creating a workspace

How to Build a Chatbot Without Coding

与讲师见面