Building Bots with Microsoft Bot Framework

Kishore Gaddam

• Éditeur: Packt Publishing
• Année: 2017
• ISBN:
  9781786469649
• Édition: 1
• Sujet: Ciencia de la computación
• Sous-sujet: Sistemas operativos

Build intelligent and smart conversational interfaces using Microsoft Bot FrameworkAbout This Book• Develop various real-world intelligent bots from scratch using Microsoft Bot Framework• Integrate your bots with most popular conversation platforms such as Skype, Slack, and Facebook Messenger• Flaunt your bot building skills in your organization by thoroughly understanding and implementing the bot development concepts such as messages (rich text and pictures), dialogs, and third-party authentication and callingWho This Book Is ForThis book is for developers who are keen on building powerful services with great and interactive bot interface. Experience with C# is needed.What You Will Learn• Set up a development environment and install all the required software to get started programming a bot• Publish a bot to Slack, Skype, and the Facebook Messenger platform• Develop a fully functional weather bot that communicates the current weather in a given city• Help your bot identify the intent of a text with the help of LUIS in order to make decisions• Integrate an API into your bot development• Build an IVR solution• Explore the concept of MicroServices and see how MicroServices can be used in bot development• Develop an IoT project, deploy it, and connect it to a botIn DetailBots help users to use the language as a UI and interact with the applications from any platform. This book teaches you how to develop real-world bots using Microsoft Bot Framework.The book starts with setting up the Microsoft Bot Framework development environment and emulator, and moves on to building the first bot using Connector and Builder SDK. Explore how to register, connect, test, and publish your bot to the Slack, Skype, and Facebook Messenger platforms.Throughout this book, you will build different types of bots from simple to complex, such as a weather bot, a natural speech and intent processing bot, an Interactive Voice Response (IVR) bot for a bank, a facial expression recognition bot, and more from scratch.These bots were designed and developed to teach you concepts such as text detection, implementing LUIS dialogs, Cortana Intelligence Services, third-party authentication, Rich Text format, Bot State Service, and microServices so you can practice working with the standard development tools such as Visual Studio, Bot Emulator, and Azure.Style and approachThis step-by-step guide takes a learn-while-doing approach, delivering the practical knowledge and experience you need to design and build real-world Bots. The concepts come to you on an as-needed basis while developing a bot so you increase your programming knowledge and experience at the same time.

Developing WeatherBot Using Dialogs and LUIS

In previous chapters, we have gone through some of the concepts involved in developing and publishing bot applications. In this chapter, we will develop a bot called WeatherBot, show you how to use LUIS in dialogs and how to use third-party APIs from a bot. This involves some additional coding efforts to develop the weather bot. We will build a weather bot that is able to understand and respond to various commands, such as What's the weather like in New York?, Get Weather in Seattle, and so on. The bot will use LUIS to identify the intent of the user and reply with the appropriate message.

Before jumping into writing code, we need to configure LUIS for WeatherBot. Here, we will go through the steps on how the user gets weather data for a given location when requested by the user. We will see how LUIS can help us make the conversation between the user and your bot in more natural language, similar to how we interact with humans.

The WeatherBot will have intelligence, which will help users to interact with it, similar to how we interact with humans (in natural language). The following are some examples:

• What will the weather be like in Ashburn?
• Get weather in Seattle
• Weather in Seattle
• Hi, what is the current weather in Ashburn?

We will achieve this with the help of Natural Language Processing (NLP) using Microsoft Cognitive Services' LUIS.

Language Understanding Intelligent Service (LUIS)

Language Understanding Intelligent Service (LUIS) is one of the services in Microsoft Cognitive Services provided by Microsoft. As mentioned earlier, natural language is a fundamental element in developing bot applications. As a result, the technology industry has seen a direct correlation between the evolution of bot platforms and NLP platforms. Although the evolution of bot technologies has been predominantly driven by messaging platform providers such as Slack or Facebook, the main advancements in NLP technologies seem to be coming from cloud platform providers such as Microsoft. As a result, to take advantage of the NLP and Natural Language Understanding (NLU) algorithms, most bot developers spend time integrating their bot applications with NLP services provided by platforms such as LUIS from Microsoft. LUIS can process natural language using pre-built or custom-trained language models.

Microsoft's LUIS is a component of the Microsoft Cognitive Services Suite that helps in creating and processing natural language models. LUIS provides a sophisticated toolset that allows developers to develop and train the platform in new conversation models. LUIS can also be used in conjunction with other text processing APIs in the Microsoft Cognitive Services Suite, such as text analytics and many other services. The LUIS platform provides a deep integration with Microsoft Bot Framework technology and can be used by other bot platforms.

Here are some of the salient features of LUIS:

Let's perform the following steps to sign up for Microsoft Cognitive Services and learn how to use LUIS:

1. Go to https://www.microsoft.com/cognitive-services and select the APIs option on the home page:

2. Under the APIs menu, select the Language Understanding option:

3. On the Language Understanding Intelligent Service (LUIS) page, click on Get started for free:

4. It will navigate to https://www.luis.ai/ , the home page of LUIS. Click on Sign in or create an account:

5. On the Sign In Options popup, select Sign in using a Microsoft account (most users):

6. It will open an OAuth flow to authenticate your Microsoft account. Once you are successfully authenticated, it will ask you to grant permissions to access your profile information; click on Yes:

7. Now LUIS will ask you to give a little more information about your country and company. After entering all the required information, click on the Continue button:

8. The following is the page where you will see all the LUIS apps, that you create:

Now we are ready to create and build LUIS models. With the help of LUIS, we can build more complex NLP models, but for the weather bot, we will use basic and pre-built features. Before building your model, you should know what an Intent is and what an Entity is.

Intents and Entities

When a user enters a sentence, LUIS will interpret it and parse out the Intent and Entities. An Intent is an action the user wants to perform, and Entities are the s...

Table des matières

• Title Page
• Copyright
• Credits
• About the Author
• About the Reviewer
• www.PacktPub.com
• Customer Feedback
• Preface
• Setting up Microsoft Bot Framework Dev Environment
• Developing Your First Bot Using the Connector and Builder SDK
• Developing WeatherBot Using Dialogs and LUIS
• Natural Speech and Intent Processing Bot Using Microsoft Cognitive Services
• Developing Bots Using LUIS Prompt Dialogs with State and Nearby Bot Using Custom APIs
• Developing an IVR Bot for a Bank Using Advanced Microsoft Bot Framework Technologies
• Intelligent Bots with Microsoft Bot Framework and Service Fabric
• Developing Intelligent Facial Expression Identification Bot for IoT Using Azure and Power BI
• Publishing a Bot to Skype, Slack, Facebook, and the GroupMe Channel