You are on page 1of 28

Chatbot - The Future of Customer Service

Hơn 30.000 loại Chatbot


mới và 6.000 ứng dụng
kích hoạt bằng giọng nói
Bot vs Chatbot vs Robot vs AI
Bot vs Chatbot vs Robot vs AI
How it work
Types of Chatbot
Types of Chatbot
Types of Chatbot
Menu/Button-Based Chatbots 📍
Types of Chatbot
Keyword Recognition-Based Chatbots ⌨️
Types of Chatbot
Contextual Chatbots
Types of Chatbot
Types of Chatbot
Why Chatbot is booming now ?

• Rise of Social messaging • NLP, NLU, NLG


Why Chatbot is booming now ?
3. Chatbot Ecosystem
Benefits
Benefits
How Artificial Intelligence Is Making Chatbots Better For
Businesses
10 Ways Enterprise Chatbot Solutions And AI
Are Changing The Workplace

1. Acting As Personal Assistants


2. Offering On-Demand Customer Support
3. Providing In-Depth Product Knowledge
4. Streamlining Business Processes
5. Empowering Your HR And IT Department
6. Manipulating Data
7. Simplifying International Payments
8. Reducing Expenses
9. Increasing Productivity
10. Helping With Business-To-Business Applications
What is the best platform to write a chatbot?

• Streebo Chatbot Builder


• TARS
• Beep Boop
• BotEngine.ai – bot framework for AI chatbot
• Botsify.com
• Bottr
• Chatfuel
• Chatkoo
• Chattypeople
• Dialogflow
• Floatbot.ai
• Flow XO
• Landbot.io | Build Conversational Websites
• Manychat
• MobileMonkey | Facebook Messenger Marketing Chatbots
• Recime | The Enterprise Bot Development Platform
• Wit.ai
• ….

FPT AI; Hekate; Hana AI; …


What is the best platform to write a
chatbot?
• Messaging Channels: Where bot deployed, such
as: Messenger, Telegram, LINE, Slack, Skype,
more.
• Enabler: All in one service that includes NLP
Engine, Building Blocks, Messaging Integrator to
create bot, sometimes it has limitation when you
want to build complex bot.
• NLP Engine: Service for extracting sentence into
bot understanding.
• Integrator: Service for third party integration
Q&A engines example
Q&A engines example

A chat bot typically has 3 things in it


• Intent ( Intention of the query asked by the user)

• Entities ( Named entities in Query like , Location names, People names,


date and etc…) #NamedEntityRecognization

• Action or Response ( the result to throw back to the user)


Q&A engines example
Step 1: Understanding the problem and preparing
conversation flows.

Step 2 : Finding Intents and entities

Step 3: Prepare the training data for getting intents and


entities.

Step 4 : Build the machine learning classifier and/or NER


(named entity recognition ) to classify the input.
Q&A engines example
Understanding the problem and preparing conversation flows
Q&A engines example

Step 2 : Finding Intents and entities


• 3 intents:
– (1) Order Pizza,
– (2) . Get price
– (3). Get offers/choices
• Entities are:
– Pizza name, → Cheese Pepproni
– Pizza size, → Large, Medium,small
– pizza type → Veg or Non veg or both
– Pizza/item quantity → 1 ,2 ,3 .
– Item name → Pepsi
– Item quantity → 1 lit,2 lit
Q&A engines example
Prepare the training data for getting intents and entities
Q&A engines example

Build the machine learning classifier