Uploaded by kyrillos.joe

Build Chatbots with Amazon AWS Innovate Dev Edition

Building Chatbots with
Amazon Lex
@awscloud
Sara Mitchell
AWS Financial Services Solutions Architect
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
#AWSInnovate
Chatbots
Good AI Chatbots Solve Some Of The Hardest Problems In
Computer Science
Learning
Language
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Perception
Problem
Solving
Reasoning
Use cases
Informational Bots
Chatbots for everyday consumer
requests
• News updates
• Weather information
• Game scores
Application Bots
Build powerful interfaces to mobile
applications
• Book tickets
• Order food
• Manage bank accounts
Enterprise Productivity Bots
Streamline enterprise work activities and
improve efficiencies
• Check sales numbers
• Marketing performance
• Inventory status
Internet of Things (IoT) Bots
Enable conversational interfaces for device
interactions
• Wearables
• Appliances
• Auto
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Developer challenges
Security
Speech
recognition
Authentication
Language
understanding
Disparate systems
Messaging
platforms
Business logic
Availability
Testing
Scale
Mobile
Conversational interfaces need to combine a large number of
sophisticated algorithms and technologies
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Amazon Lex – Features
Text and speech language understanding: powered by the
same technology as Amazon Alexa
Deployment to chat services
(Web/Mobile Apps, Facebook, Kik, Slack, Twilio SMS)
Designed for builders: efficient and intuitive tools to build
conversations; scales automatically
@
Versioning and alias support
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Amazon Lex Bots – key concepts
BookHotel
Intents
An intent performs an action in
response to natural language user input
Utterances
Spoken or typed phrases that invoke
your intent
Slots
Slots are input data required to fulfill
the intent
Fulfillment
Fulfillment mechanism for your intent
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
“Book a hotel”
Book
a
in
“Book a hotel in
NYC”
Hotel booking
hotel
New York City
NYC
Automatic speech
recognition
Natural language
understanding
Hotel Booking
“Your hotel is booked for
Nov 30th”
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Utterances
City
New York City
Check in
Nov 30th
Check out
Dec 2nd
Amazon Polly
Intent/slot
Model
“Can I go ahead
with the booking?
Confirmation: “Your hotel is
booked for Nov 30th”
Utterances
I’d like to book a hotel
I want to make my hotel reservations
Can you help me book my hotel?
I want to book a hotel in New York City
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Slots
Slot
Type
Values
Destination
City
New York City, Seattle, London …
Check in
Date
Valid dates
Check out
Date
Valid dates
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Slot elicitation
I’d like to book a hotel
Sure, what city do you want to book?
City
New York City
New York City
What date do you check in?
Nov 30th
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Check in
11/30/2017
Amazon Lex - Technology
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Create the bot
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Create and edit a slot type
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Business logic provided by Lambda
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
My Lex Bot
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Designing conversational
interfaces
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Conversational Interfaces
•
Don’t drone on
•
Use natural language– think about how you’d say something if you were
face to face
•
•
•
•
•
Taking turns
Maintaining context
Manage colloquial phrases
Variety – different ways of saying the same thing
Handling the unexpected
•
•
•
People don’t respond in the ‘correct’ way but don’t want to repeat themselves
Error handling should be brief and focused on getting conversation back on
track
Offer suggestions – again to get the conversation moving forward and avoiding
dead ends
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
What about if I could call my Lex bot?
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Amazon Connect
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Amazon Connect
Easy to use, cloud-based contact center solution that scales to
support businesses of any size
With Tools That Grow With Your Needs
Skills-based routing
[Automatic Call Distribution (ACD)]
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Call
recording
Real time and
historical analytics
High-quality
voice capability
Contact Flow Engine – Customer Experience
Incoming
customer
call
CRM
content
Data
Dip
DYNAMIC
Answer customer
questions before they
are even asked
Hi Nikki Wolf,
I see your flight
was cancelled
today. How can
I help you?
Can you
please
rebook me
for the
same flight
tomorrow?
PERSONAL
Contact flows adapt
on a per customer
basis
Ok, you are now
booked for a
9:00AM departure
tomorrow out of
San Francisco,
arriving in Seattle
at 11:45AM.
Great
Thank you!
NATURAL
Amazon Lex Chatbots
use the same technology
that powers Alexa
Extend your chatbot with easy Integrations
OPEN PLATFORM
Customer
Databases
Business
Intelligence
Your
S3 Storage
Call
Recordings
AWS
Lambda
Contact
Flows
Agent Data
Metrics
CRM
Contact
Control Panel
Your Data
Warehouse
Workforce
Management
Summary
•
•
•
•
•
Conversational interfaces built with Amazon Lex
Business logic built in AWS Lambda
Voices provided by Amazon Polly
Call Centre running on Amazon Connect
Extend it, for example:
•
•
•
•
Integrate the bot with text based services such as Slack, Twilio
Call Recordings Stored in S3 – audit and analysis
Integrate with CRM tools
Click to dial from Salesforce
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Thank you!
@awscloud
#AWSInnovate
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Useful reading
https://aws.amazon.com/blogs/machine-learning/building-better-bots/
https://aws.amazon.com/blogs/machine-learning/building-better-botspart-2/
https://aws.amazon.com/blogs/machine-learning/greetings-visitorengage-your-web-users-with-amazon-lex/
https://aws.amazon.com/blogs/aws/new-amazon-connect-andamazon-lex-integration/
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.