AWS Public Sector Blog
Building a multi-channel Q&A chatbot at Saint Louis University using the open source QnABot
Students have questions. Their schools have the answers. A multi-modal chatbot delivers the missing link, providing a way for students to ask questions and get access to institutions’ answers.
How does it work? Build an Amazon Alexa skill so that students can ask Alexa for answers on a variety of devices: Amazon Echo Dots and Echo Shows, FireTVs, or Fire tablets. Or add chat bots to your website, with a web user interface (UI) that allows students to type or speak their questions on their laptops or mobile phones. Or create an SMS bot so students can text questions and instantly get answers by reply text when they’re on the go.
Or, like Saint Louis University, use the open source QnABot on Amazon Web Services (AWS) to create a chatbot that supports all these modes and more. QnABot features can simplify bot deployment and administration for non-technical users. For example, you can:
- Deploy into your AWS account using CloudFormation, at the push of a button
- Use the Content Designer UI to add and edit content to make your bot smarter
- Optionally customize the answers by mode; for example use buttons and markdown for a rich web experience, and SSML for a rich voice experience
- Extend the bot using Lambda hooks to fetch dynamic answers from external sources
- Capture user feedback with “Thumbs Up” and “Thumbs Down” responses
- See what questions people are asking, making it easy to add new features and functionality as your users demand them
- Direct users to your website or helpdesk when the bot doesn’t have the answer to a question
The SLU story
Last year, Saint Louis University (SLU) was the first university to bring virtual personal assistants into every student residence, placing 2,300 Amazon Echo dots equipped with the “AskSLU” skill into residence hall rooms on campus. The private skill, managed by Alexa for Business, was able to answer more than 100 questions, including questions about Billiken athletics games, concerts and major speakers, student events and organizations, and service and mission opportunities, along with many other aspects of student life. The skill was designed to provide students with information to quickly acclimate to campus life and to help them connect to their community. David Hakanson, the University vice president, CIO and chief innovation officer said they went “all in” with Alexa because, “every minute we can save our students from having to search for the information they need online is another minute that they can spend gaining the most from their college experience.”
When SLU launched the skill, they developed a mechanism similar to the “What’s New with Alexa” newsletter to share monthly updates and functionality with students and to solicit student feedback about the skill. Students shared they wanted more information about dining services, athletics information, and bus schedules, and would like to be able to communicate with the skill from outside their dorm rooms, using text messages or the website.
Over the summer, SLU worked with AWS Professional Services to move their private Alexa for Business skill over to the open source QnABot project. Today, students and community members are able to interact with the newly launched Amazon Alexa “Saint Louis University” skill on Amazon Echo devices managed by Alexa for Business in student residence hall rooms, or on any Alexa capable device (Amazon Echo, FireTV, Fire tablets, etc.) that enabled the skill from the skill store. The question and answer experience provided by Alexa has now been extended so users can interact with the bot on the SLU website, or can access it while on-the-go through SMS text messaging. SLU leveraged the QnABot framework to develop an experience that end users can access on three different channels, with answers customized to the channel used.
With the new QnABot-powered bot, the SLU team can now easily see which questions are most popular and which answers elicited negative feedback. With the Content Designer UI, they can quickly fix problems and continuously make the bot smarter by tailoring responses and adding new questions to the set of queries that the bot knows how to answer.
The new version of the chatbot also now integrates with a variety of backend information systems, such as Sodexo food services, a database of building hours, transport schedule systems, athletics information, and more. The “Saint Louis University” skill can now provide dynamic answers to questions like, “What are the gluten-free lunch options at the Dining Hall today?,” “When does the next bus arrive at bus stop Salus Center?,” “Is the library open on Saturday?,” “When is the next basketball game?,” and more.
Create your own QnABot & learn more
To get started, read the QnABot blog post and follow the tutorial. If, like SLU, you want to add QnABot to your website, try our companion Chatbot UI project. Decide how you’d like to proceed. Options include:
- Deploy the QnABot and Chatbot UI yourself (self-serve), using the project “as-is.”
- Make your own customizations and enhancements to the open source code.
- Follow SLU’s example and contact AWS Professional Services for expert help to customize and enhance QnABot, integrate with external information sources, add support for popular messaging apps or integrate into your call center, or enable user authentication and support personalized answers.
To collaborate with like-minded institutions to share experiences and accelerate development of new features and integrations with commonly used backend systems, contact Kyle Collins, Assistant Vice President – Technology Transformation at email@example.com.
Meet Kyle Collins from SLU at AWS Re:Invent 2019 to hear more about the chatbot project and ask questions at session WPS203 “AWS Educate: Innovation in Education.” Also at Re:Invent – join a hands-on workshop to try out the latest version of QnABot during session AIM302 “Create a Q&A bot with Amazon Lex and Amazon Alexa.”