Monday, December 6Digital Marketing Journals

Why you should consider Microsoft Bot Framework for your next AI Chatbot | by ximnet | Nov, 2021


Building an AI Chatbot has never been easier

ximnet

Microsoft Bot Framework is a framework for building enterprise-grade conversational AI experiences. It is hosted on Microsoft Azure’s cloud infrastructure. With this framework, we have launched a few virtual assistants such as Meva for PETRONAS Dagangan Berhad’s MyMesra, Skye for Kuala Lumpur Convention Centre, and Nadia for Affin Hwang Asset Management. In this article, I’m going to share why you should consider Microsoft Bot Framework for your next AI Chatbot as a developer.

Skye — Virtual Assistant for Kuala Lumpur Convention Centre

First, this bot framework is very easy to start. We can have a simple Q&A bot using QnA Maker in less than an hour without coding at all. You can upload an excel file of questions and answers or provide a webpage. You can refer to this link to get your first bot up and running. Some bots may continue as a simple QnA bot, and some may evolve to be a more complex bot.

A bot may need to be very flexible, or complex based on clients’ business requirements. Here, we can use the Microsoft Bot Framework SDK to build our complex bot dialogs and integrations. We can program complex nested dialogs to run product recommender, for example. You can also program the bot to guide users to give feedback or make an appointment.

Lubricant Recommender for Motorcycle in MeVA

Within the bot dialogs, we can also integrate our bot with Microsoft LUIS which is a Cognitive Service solution. LUIS allows us to get a user’s intents and entities to make our bot smarter. For example, if a user’s intent is to submit a complain, we can use LUIS to get their name, email, and phone numbers accurately. LUIS also can get dates, measurements and also custom entities using regular expressions such as your countries’ identity number.

Since the bot can be hand coded, we can also connect to any REST API. For example, we can connect the Azure Text Translator to translate Malay language into English to be processed by LUIS. This is because there are languages that are not supported by LUIS natively. We can also connect to our other apps to push or get data.

1. How Conversational AI can Automate Customer Service

2. Automated vs Live Chats: What will the Future of Customer Service Look Like?

3. Chatbots As Medical Assistants In COVID-19 Pandemic

4. Chatbot Vs. Intelligent Virtual Assistant — What’s the difference & Why Care?

Even though this bot framework is built by Microsoft, we can use 4 types of different programming languages ranging from C#, Java, JavaScript to Python to build the bot. We do not need to learn another language if we already know one of those programming languages. There are also bot project templates for Visual Studio to help us to get started. The templates come with samples on how to do certain tasks in a chatbot.

For testing the bot, we can use Bot Framework Emulator which is installed onto our machine. This tool is a conversation interface to debug the bot locally with our development tool. It is fully integrated with Visual Studio, and it makes testing and debugging work a lot easier. For example, we can set breakpoints in the code and do live debugging.

Microsoft Bot Framework is also ready for integration with XTOPIA. By integrating with XTOPIA, your end users can update the bot’s responses easily and enhance your website with bot interaction. The end users can also make your bot better with XTOPIA codeless workflow by getting data from internal and external sources.

MeVA’s integration with Petrol Station’s location API using XTOPIA.

That’s all for now. We will share more about Microsoft LUIS and bot development in the future. You can find out more about Microsoft Bot Framework here and XTOPIA platform here.

Leave a Reply