Artificial intelligence for minors by minors

July 22, 2019 devadvin

My name is Jazmin Carrillo Morales, and I am an incoming senior at Woodside High School in California. I’m spending eight weeks as a summer intern at IBM’s Watson West office in San Francisco. I had no prior experience working with AI technology, so I knew this internship would be a great learning experience. I am glad to be an IBM intern because I have the ability to learn more about different careers in the company. I am not yet certain what career I want to pursue, which is why I am trying to learn as much as possible and get a feel of what it is like to work for a large technology company.

My name is Miguel Tostado, and I am an incoming senior at Menlo Atherton High School. I am interested in computer science and mechanical engineering, but I’m not yet sure what I want to study, which is why I am excited to be an intern at IBM. At IBM, I want to gain experience and explore the various careers IBM has to offer. In my time here, I have worked on a couple of projects, one being a chatbot, which we will explain on how we created.

We wanted to experiment with IBM Watson® and our manager, Peter Moskovits, mentioned that we could create a chatbot. We decided to create our chatbot on basketball because it is a topic we are both interested in and want to learn more about. We enjoy watching the Golden State Warriors. This past season, we were excited when the Warriors made it to the finals, but we were disappointed when they lost to the Toronto Raptors. We are basketball fanatics, so we created a chatbot that can answer questions about the 2018-19 NBA season. We managed to create a full-functioning chatbot with IBM Watson using resources like YouTube and Google to help guide us in creating it.

Steps

  1. The first step is to search for IBM chatbot.

    alt

  2. Click on Get started free to start the chatbot.

    alt

  3. Use your IBM Cloud login information to access IBM Watson.

    alt

If you don’t have an IBM ID, you can easily create one for free.

  1. Click on Create a skill.

    alt

  2. Click on Create a skill once again. Disregard the basketball skill on the screen since that is the one we created.

    alt

  3. Once you have named and created the skill, you are ready to start your chatbot. There are three key components that make up the chatbot: intents, entities, and dialog.

Intents

Intents are customer goals expressed in user input. Determine the types of questions and requests you want your assistant to understand and provide example utterances for Watson to learn from. You can add prebuilt intents from content catalogs, such as general greetings. (See Intents Explained by an IBMer for more information.)

alt

Entities

Entities clarify a customer’s purpose by allowing the assistant to recognize synonyms. When Watson recognizes the synonym input into the system, the assistant will choose the best way to respond to the user. (See Entities Explained by an IBMer for more information.)

alt

Dialog

Design a dialog flow for the assistant to follow while it interacts with customers. A good flow triggers appropriate and useful responses based on the intents and entities detected in the user input. (See Dialog Explained by an IBMer for more information.)

alt
alt
alt

Once you have created a node under Dialog, you can test your assistant and ask a question that can be answered with the information you have entered in the node.

alt

Check out our Basketball chatbot and give it a try. Some questions you can ask are:

  • Who won the 2019 championship?
  • Who won MVP this season?
  • How long is a basketball game?
  • Who is Stephen Curry?
  • Who is MJ?

Previous Article
Congratulations to the Apache OpenWhisk community on its graduation
Congratulations to the Apache OpenWhisk community on its graduation

Apache OpenWhisk went from being an Incubating Project to being a Top-Level Project at the Apache Software ...

Next Article
Developer relations down the stack
Developer relations down the stack

An interview with Marek Sadowski, an expert in containers, Kubernetes, and server-side Swift who started ou...

×

Want our latest news? Subscribe to our blog!

Last Name
First Name
Thank you!
Error - something went wrong!