Learn AI with a No-Code Bot That Understands You

Table of Contents

• Introduction
• What is Build 2020?
• Code of Conduct for Build 2020
• Creating an AI without Writing Code
• Getting Started with the LUIS Platform
• Designing the Schema
• Training the AI Model
• Using Pre-built Domains in LUIS
• Understanding Intents and Entities
• Testing and Publishing the AI Model

🎇 Build 2020: Creating an AI without Writing Code 🎇

Welcome to the grand finale of Build 2020! I hope You've enjoyed the last 47 and a half hours of incredible Sessions. If you're still here, congratulations! Before we proceed, I want to emphasize the importance of kindness and respect in the comments and in our interactions as a community. Together, we can Create amazing things.

Now, without further ado, let's dive into the fascinating world of artificial intelligence. In this session, we will be featuring the one and only Instafluff on Twitch, who will Show us how to build an AI without writing any code. Yes, you heard that right – creating an AI without coding! So, sit back, relax, and let's get started on this exciting Journey.

🌟 Getting Started with LUIS Platform

To begin, let's talk about how you can get started with the LUIS platform. LUIS, or Language Understanding Intelligent Service, is a powerful tool that allows you to train your AI model to understand natural language. To start creating your own app, simply visit the LUIS Website at www.luis.ai.

Once you're on the website, you can create a new app for conversation. LUIS is a machine learning platform that learns from examples. You can train it with the specific knowledge you want it to learn. It's like building a brain for your AI.

📚 Designing the Schema

Before we dive into the training process, let's talk about designing the schema. The schema defines the structure of your AI model. It specifies what kind of intents and entities your AI should be able to recognize.

For example, if you're creating a weather app, you might have intents like "QueryWeather" and entities like "City" and "Temperature". By defining these intents and entities, you are essentially teaching your AI what kind of information it needs to extract from user input.

Once you have the schema figured out, you can start building your model. This is where you're coming up with all the intents, entities, and example sentences that your AI will learn from. It's like training your AI to understand and respond to different user inputs.

🧠 Training the AI Model

Training your AI model is a crucial step in the process. It's where you teach your AI to recognize Patterns and understand user inputs. The more examples you provide, the better your AI will become at understanding different variations of user queries.

Let's take the example of a weather app. You can train your AI model with a wide range of example sentences related to weather queries. For instance, "What's the weather in Seattle?" or "Tell me the temperature in Celsius." By providing a diverse set of examples, you're enabling your AI to handle different user inputs effectively.

🌐 Using Pre-built Domains in LUIS

One of the cool features of LUIS is the ability to use pre-built domains. These domains provide a set of pre-defined intents and entities that help you recognize specific categories of user queries. For example, if you want to create a booking app, LUIS offers pre-built domains for actions like "PurchaseAirlineTicket", "RentCar", and "ReserveHotel".

By leveraging these pre-built domains, you can save time and effort in defining intents and entities from scratch. It's like having a head start in training your AI model to understand specific use cases. You can even customize these pre-built domains to suit your app's needs.

💡 Understanding Intents and Entities

Intents and entities are the building blocks of your AI model. Intents represent the goals or actions a user wants to perform, while entities are the specific pieces of information that your AI needs to extract from user inputs.

For example, in our weather app, "QueryWeather" is an intent, and "City" and "Temperature" are entities. Intents help your AI understand what the user wants, while entities provide the necessary Context for fulfilling that request.

By accurately defining and training your intents and entities, you can ensure that your AI model delivers accurate and Meaningful responses to user queries.

🚀 Testing and Publishing the AI Model

Once you have trained your AI model, it's time to test and publish it. Testing allows you to verify that your model is correctly recognizing intents and entities and providing accurate responses. You can use the test feature in LUIS to simulate user interactions and see how well your model performs.

If everything looks good, you can proceed to publish your AI model. Publishing makes your model available as an endpoint that you can call from your applications or services. You can integrate it into any web or mobile app using HTTP calls or SDKs provided by Azure.

With your AI model published, you can now start reaping the benefits of your hard work. Whether you're building a chatbot, a voice assistant, or any other AI-powered application, your model is ready to analyze and understand user inputs.

🌟 Highlights 🌟

• LUIS allows you to create AI models without writing code
• Designing a schema is crucial for defining intents and entities
• Pre-built domains in LUIS can save time and effort in training your AI model
• Accurately defining and training intents and entities improves the accuracy of your AI model
• Testing and publishing your AI model ensures its reliability and availability

🙋‍♀️ FAQ

Q: Can I use LUIS with languages other than English? A: Yes, LUIS supports multiple languages. You can create different versions of your app to handle different languages.

Q: Is there a limit to the number of intents and entities I can train in LUIS? A: While there is no hard limit, it is generally recommended to have a similar number of training examples for each intent or entity to achieve better accuracy.

Q: Can I keep my AI model trained in LUIS or do I need to train it regularly? A: LUIS allows you to update and retrain your AI model as needed. You can keep your model trained and continually improve its performance.

Q: Can I use LUIS with Azure Bot Framework? A: Yes, LUIS can be integrated with Azure Bot Framework to create powerful chatbots. They complement each other in providing natural language understanding and chatbot capabilities.

Q: Are there other Azure services that work well with LUIS? A: Yes, Azure offers a range of cognitive services that can complement LUIS, including speech recognition and translation services. You can enhance your AI model with these services to provide a more immersive experience to users.

Resources

• LUIS
• Azure Bot Framework
• Azure Cognitive Services