AI & CHATBOTS
Automating customer relation
20 people in 2017
7 people in R&D
13 people in the Account & Project team
100% excellency
2
Who are we?
Introduction
Defining the concept of AI
✓ Data-powered
✓ Algorithmic science
✓ Machine Learning
Introduction
The notion of chatbots is nothing new
✓ Eliza in 1964
✓ Cleverbot in 1997
✓ Bots for Messenger in April 2016
Plan
I. Chatbots: state of the art
II. What does a chatbot project entail?
III. In the future
Chatbots: State of the art
Plan
1. Intro
2. AI and current Chatbots
3. Hopes vs reality
4. Natural Language Understanding
5. Our approach
“30 or 40% of our clients’ messages are recurrent and could
be partly automated. ”
Axa France
In China, a brand will
create a bot on WeChat
before creating a website.
3 billion
users
An indeniable trend
An indeniable trend
Conversation Platforms Online Chat
Messaging apps (FB
Messenger, Slack...)
Intercom
iAdvize (Q2 2017)
Zendesk (Q2 2017)
SMS Vocal
Twilio
Phone
Alexa / Google
Assistant (Q3 2017)
Current integrations
Problem
AI and current Chatbots
An undeniable trend
April 2016: 30 000 bots created in 3 months but…
« Bots right now are in the trough of despair. To industry
observers, it feels like they are overhyped and under-delivering. »
Greg John, CEO of Burner
Current chatbot technology is nothing new. It becomes
interesting when chatbots meet AI.
Hopes vs Reality
Understanding the limits of AI Chatbots
✓ Bots are not yet intelligent
(language, context)
✓ Questions need to be predicted
✓ Answers need to be written in
advance
✓ Complicated features take time
to develop
Some bad examples
Х Tay: Microsoft’s error in 2016
Х M: Facebook’s perfect bot
Natural Language Understanding
The basics
To build a chatbot able to converse with human, you need NLU technology.
✓ Data-powered
✓ Detecting intentions
✓ Focused on keywords and trigger
words
✓ Understanding words in contexts
Our approach
Data-based chatbots
✓ We analyze historical datasets
✓ We detect FAQs
✓ We map those questions and detect intentions
✓ We input this data in the bot
Example
16
17
Bot Admin &
Analytics
Your Data
Pull Data
Push API
Your Data
Push Data
Webhook
Trigger actions and
answers (Conversation
Management platform)
Intent detection
(Clustaar Deep Query)
User input
Integration with data
Log analysis
✓ After a few days in production, then once a month,
unanswered questions go through our algorithm Deep
Query©.
✓ New “intents” are identified and made accessible to the
bot. They need to be then associated with the
appropriate actions.
✓ The bot becomes can handle more and more cases on
its own.
What does a chatbot project entail?
Plan
1. How to envision a bot project?
2. Use cases
3. How to make it great?
Envision a chatbot project
Designing a new user experience
✓ Define the objectives of the bot
✓ Adapt to the target
✓ Recreate user habit
✓ Transform them into conversation
✓ Imagine client reactions
Possible use cases
All recurrent interactions
Internal
(ex: Nexity)
- HR
- Knowledge
management
- Search in databases
- FAQ
Client acquisition
(ex: Cortex)
- Integrated to a
website
- Automatic lead
generation
- FAQ
- Available 24/7
Service
(ex: Hachette)
- Playful features
- Geolocalisation / Store
locator
- Promotional offers or
news Push
- FAQ
Customer service
(ex: 20 Minutes)
- Integrated to a website
or Messenger
- Automatic FAQ in
Natural Language
- Available 24/7
Client acquisition Customer service
How to make it great
Put a smile on the client’s face
✓ Smalltalk: simulating human interaction and setting a tone
✓ Fallback: building a fluid conversation
✓Playful features: jokes, quiz…
Find the detail that will make the user say « thank you! »
Data import Bot training Put the bot online Run, manage & improve
Internal sources (FAQ,
conversations) and external
sources (Google Queries)
Intent detection
Connexion with internal data
Writing scenarios and
answers
UX & integration
Publication
Fine tuning
Machine learning
improvement & Analytics
Improving response
scenarios
25
Project phases
In the future
Plan
1. Outstanding Conversation
2. Understanding emotions
3. Personnalized insurance
Outstanding conversation
Flawless automated client interaction
✓ Understanding complex questions
✓ Automatically generated answers
✓ 100% automated customer care
✓No interface ?
No interface
All vocal conversation
Understanding emotions
Mesuring client satisfaction automatically
✓ Live sentiment analysis
✓ Emotionaly appropriate response
✓ Automatic report on client satisfaction
Personnalized insurance
Data collection to a new level
✓ Perfect perception of clients’ profiles
✓ Automatic daily personnalization
Personnalized insurance
It’s all about data
Personnalized insurance
It’s all about data
www.clustaar.com/en
Mail
nicolas@clustaar.com
philippe@clustaar.com
Phone
Nicolas Chollet
+33 (0) 6 51 42 79 05
Philippe Duhamel
+33 (0) 6 84 33 76 23
Offices
28 rue du faubourg
poissonnière
75010, Paris
France

Clustaar chatbot intervention for Crédit Agricole 19/05/2017

  • 1.
    AI & CHATBOTS Automatingcustomer relation
  • 2.
    20 people in2017 7 people in R&D 13 people in the Account & Project team 100% excellency 2 Who are we?
  • 3.
    Introduction Defining the conceptof AI ✓ Data-powered ✓ Algorithmic science ✓ Machine Learning
  • 4.
    Introduction The notion ofchatbots is nothing new ✓ Eliza in 1964 ✓ Cleverbot in 1997 ✓ Bots for Messenger in April 2016
  • 5.
    Plan I. Chatbots: stateof the art II. What does a chatbot project entail? III. In the future
  • 6.
  • 7.
    Plan 1. Intro 2. AIand current Chatbots 3. Hopes vs reality 4. Natural Language Understanding 5. Our approach
  • 8.
    “30 or 40%of our clients’ messages are recurrent and could be partly automated. ” Axa France In China, a brand will create a bot on WeChat before creating a website. 3 billion users An indeniable trend
  • 9.
  • 10.
    Conversation Platforms OnlineChat Messaging apps (FB Messenger, Slack...) Intercom iAdvize (Q2 2017) Zendesk (Q2 2017) SMS Vocal Twilio Phone Alexa / Google Assistant (Q3 2017) Current integrations
  • 11.
  • 12.
    AI and currentChatbots An undeniable trend April 2016: 30 000 bots created in 3 months but… « Bots right now are in the trough of despair. To industry observers, it feels like they are overhyped and under-delivering. » Greg John, CEO of Burner Current chatbot technology is nothing new. It becomes interesting when chatbots meet AI.
  • 13.
    Hopes vs Reality Understandingthe limits of AI Chatbots ✓ Bots are not yet intelligent (language, context) ✓ Questions need to be predicted ✓ Answers need to be written in advance ✓ Complicated features take time to develop Some bad examples Х Tay: Microsoft’s error in 2016 Х M: Facebook’s perfect bot
  • 14.
    Natural Language Understanding Thebasics To build a chatbot able to converse with human, you need NLU technology. ✓ Data-powered ✓ Detecting intentions ✓ Focused on keywords and trigger words ✓ Understanding words in contexts
  • 15.
    Our approach Data-based chatbots ✓We analyze historical datasets ✓ We detect FAQs ✓ We map those questions and detect intentions ✓ We input this data in the bot
  • 16.
  • 17.
    17 Bot Admin & Analytics YourData Pull Data Push API Your Data Push Data Webhook Trigger actions and answers (Conversation Management platform) Intent detection (Clustaar Deep Query) User input Integration with data
  • 18.
    Log analysis ✓ Aftera few days in production, then once a month, unanswered questions go through our algorithm Deep Query©. ✓ New “intents” are identified and made accessible to the bot. They need to be then associated with the appropriate actions. ✓ The bot becomes can handle more and more cases on its own.
  • 19.
    What does achatbot project entail?
  • 20.
    Plan 1. How toenvision a bot project? 2. Use cases 3. How to make it great?
  • 21.
    Envision a chatbotproject Designing a new user experience ✓ Define the objectives of the bot ✓ Adapt to the target ✓ Recreate user habit ✓ Transform them into conversation ✓ Imagine client reactions
  • 22.
    Possible use cases Allrecurrent interactions Internal (ex: Nexity) - HR - Knowledge management - Search in databases - FAQ Client acquisition (ex: Cortex) - Integrated to a website - Automatic lead generation - FAQ - Available 24/7 Service (ex: Hachette) - Playful features - Geolocalisation / Store locator - Promotional offers or news Push - FAQ Customer service (ex: 20 Minutes) - Integrated to a website or Messenger - Automatic FAQ in Natural Language - Available 24/7
  • 23.
  • 24.
    How to makeit great Put a smile on the client’s face ✓ Smalltalk: simulating human interaction and setting a tone ✓ Fallback: building a fluid conversation ✓Playful features: jokes, quiz… Find the detail that will make the user say « thank you! »
  • 25.
    Data import Bottraining Put the bot online Run, manage & improve Internal sources (FAQ, conversations) and external sources (Google Queries) Intent detection Connexion with internal data Writing scenarios and answers UX & integration Publication Fine tuning Machine learning improvement & Analytics Improving response scenarios 25 Project phases
  • 26.
  • 27.
    Plan 1. Outstanding Conversation 2.Understanding emotions 3. Personnalized insurance
  • 28.
    Outstanding conversation Flawless automatedclient interaction ✓ Understanding complex questions ✓ Automatically generated answers ✓ 100% automated customer care ✓No interface ?
  • 29.
  • 30.
    Understanding emotions Mesuring clientsatisfaction automatically ✓ Live sentiment analysis ✓ Emotionaly appropriate response ✓ Automatic report on client satisfaction
  • 31.
    Personnalized insurance Data collectionto a new level ✓ Perfect perception of clients’ profiles ✓ Automatic daily personnalization
  • 32.
  • 33.
  • 34.
    www.clustaar.com/en Mail nicolas@clustaar.com philippe@clustaar.com Phone Nicolas Chollet +33 (0)6 51 42 79 05 Philippe Duhamel +33 (0) 6 84 33 76 23 Offices 28 rue du faubourg poissonnière 75010, Paris France