Chatbots, Frameworks and Tools, Thought Leadership

Chatbot Intent and Entities

Denis Rothman, Artificial Intelligence Author, Speaker and Instructor, partnered with Discover.Bot to speak on chatbot intent and entities, walking through a demonstration of how to use Discover.Bot to promote your professional or personal brand in this video.

By Denis Rothman
August 29, 2022

Chatbot Intent and Entities

Denis Rothman, Artificial Intelligence Author, Speaker and Instructor, partnered with Discover.Bot to speak on chatbot intent and entities, walking through a demonstration of how to use Discover.Bot to promote your professional or personal brand in this video.

Find the transcript of the recording below to follow along with the video!

Denis Rothman on Biteable.

Introduction
Today, I am going to show you why it's important to use Discover.bot to promote your personal and business image, which will help you share your ideas and expand your network. Now, let me dig into this and let me share my screen and show you what I mean.

Google Dialogflow

  • Google Dialogflow offers a very easy to use system where you're going to have Intents. The intent could be to explain algorithms, explain something about blockchains, or just to say something about one of the chapters in a book.
  • Now, if you look at here, let's click on algorithms to see how this works. Now this is very interesting for you, because in the old days we needed to write a lot of sentences so that the system would know what algorithms are.
  • Today, we don't need to do that anymore. We could just write one sentence. You're looking at a bot that's been running for quite some time. So you can say, '' What are algorithms?'' Or you could do something related to a field.
  • Suppose you're working with a doctor. You can say, ''Oh, what are the symptoms you're feeling?'' And then you can create a little dialogue with an expert system and the system will find a diagnosis.
  • Or if you're in manufacturing, you can create questions like, ''How is the production going?'' And you can connect the system to production and get answers.

Entities

  • So then you have intents, what you intend to talk about and do, and you have entities where you're just going to name all these entities you want to talk about, algorithms, blog, directions, networks.
  • So you work on that and at one point, you're going to have a pretty big dialogue. Now the best way to do that is write it on a piece of paper, then maybe represent it in a graph, talk to people about it. What we're doing here is creating an agent. It's called an agent. And this agent, once you're on discovery with a domain of your own, will talk to people 24 hours a day. That's what's interesting about it. Even when you're sleeping, people will be able to talk to you. And that's important. In the days that are ahead of us, people are just impatient to get answers and your system will work 24/7.

Testing

  • So now let's have a look at a test, because you can test it. You can test the system. So like right here, I'm going to say... I'm going to ask the system something. Say talk to artificial intelligence by example. Just something, we have to start somewhere. And you can see that the system is saying, ''Alright. Getting the test version of Artificial Intelligence by Example. Greetings! I'm here to assist you with artificial intelligence.'' Say, ''Chapter one, chapter eight.''.
  • In fact, you can say what you want. For example, I can click on the mic. I say, ''What is the Markov Decision Process?'' What is the Markov Decision Process? So now it's looking for an answer and you can see how intelligent this system is, because in fact, if you noticed, I clicked on the mic too early and the system was just picking... But it's become so powerful that it picked up the core of what I wanted. ''The Markov Decision Process is a reinforcement learning algorithm based on a random process. If you wish to know more, ask me about the Bellman equation.'' So let's ask about that and we'll stop there. Stop the testing.
  • So, ''What is the Bellman equation?'' Okay, now it's looking for the answer, and here you get it. ''It's based on state equals present state plus the highest reward.'' You get the answer. So now you can see what I'm talking about. You can build absolutely any dialogue you want, and you can see how intelligent it is by just finding things that we didn't... We were confused when we talked about it, but it will find the core topic. And it's easy to create. Now you can do it with Dialogflow, Pandora, other systems.
  • Discovery

    • Now you go to discovery. You're ready. You have your content. Remember it's important, what do you want to talk about? You created it, you created the flow, and now you want to go to a bot. You want to go to discovery and you want to get a domain.,
    • So you go to your domain and you say, ''Well, I want my AI domain and I want to create a bot with it.'' So now you're going to ask for that domain and you're going to get to Amazon Registry.
    • Now what's essential to understand is you're on Amazon. You're on a neutral platform, it's not a social media platform, which means there's no color to it. There's no political color or anything. It's just business. So Amazon Registry is a nice place. And when you promote it, you can say, ''Well, my system is on Amazon.'' Now that's important because everyone knows Amazon.
    • So all you have to do, once you created your dialogue, is just choose the platform used. I use Dialogflow in this video, but you can use another platform. You could use PandoraBots. There are several out there. Then you go to your Console on Google-proof and you get a key. I'm not going to show you my key.
    • So here you, once you have it, well you enter the bot name and you enter the client access token, you validate it. And there you go. You'll have a system that will be live on this domain. Exactly as I showed you when we were in Dialogflow and testing it.

    Check out Discover.Bot
    So that means at this point, let me sum it up, let me sum it up. At one point, you have a lot to say, either on a personal basis or professional basis, you can't be available 24/7. But you can represent all your ideas in an agent in Dialogflow or another system.

    You can create all of these question answers and create a tree, a dialogue tree. And once you have it, you can deploy it on Discovery, with your own bot domain, which will be yours. And you can share it on social media, on your website, anywhere. So you'll be promoting yourself 24/7 using Amazon tools on Discovery and you will expand your network. There's no doubt about that.

    So I hope you enjoyed the video. And what's your next step? Go to Discovery and continue the journey!