CREATING CHATBOTS

WITH GOOGLE
DIALOGFLOW
1. Chatbot Fundamentals
1.1 What is a chatbot?
A chatbot is a software application that mimics conversation with a human
in natural languages through various platforms like messaging, websites,
mobiles etc. The chatbot responds by identifying the intent of the
conversation and then responding accordingly.

Image Courtesy: https://expertsystem.com/chatbot/

Broadly, chatbots can be classified into two categories:

a) Rule-Based Chatbots: This is an extremely fundamental type of chatbot
which generally works on simple ‘if-else’ constructs. It can respond to
only simple predefined queries. The performance of this application
highly depends on the programming skills of the developer.
b) Chatbots with Natural Language Understanding: At the core, it has a
language processing and understanding model with pre-trained instances
using Deep Learning. It can communicate through both text and speech.

1.2 Uses of chatbots:

a) Can be used to answer FAQs.
b) Can be used for grievance handling.
c) Internal organizational automation.
d) To do the flight, hotel, appointment bookings etc.
e) Can guide customers to buy the correct product by answering their
questions
f) Can be used for Customer Relationship Management.

1.3 Advantages of Chatbots

a) 24*7 customer support.
b) Uniform customer experience.
c) Cost-efficient.
d) Build once and deploy everywhere.
e) Integration with various channels and platforms
f) Better monitoring and insight generation.
1.4 Frameworks Present in the Market:
 a) Google Dialogflow: Dialogflow is a natural language understanding
platform that makes it easy to design and integrate a conversational user
interface into your mobile app, web application, device, bot, interactive
voice response system, and so on.
b) Microsoft Bot Builder with LUIS: Azure Bot Service enables you to
build intelligent, enterprise-grade bots with ownership and control of
your data. 
c) Amazon Lex: Amazon Lex is a service for building conversational
interfaces into any application using voice and text. Amazon Lex
provides the advanced deep learning functionalities of automatic speech
recognition (ASR) for converting speech to text, and natural language
understanding (NLU) to recognize the intent of the text, to enable you to
build applications with highly engaging user experiences and lifelike
conversational interactions.
d) RASA: Rasa provides infrastructure & tools necessary for high-
performing, resilient, proprietary contextual assistants that work.
e) Wit.ai (Facebook): Wit.ai makes it easy for developers to build
applications and devices that you can talk or text to.

2. Google Dialogflow
2.1 Introduction:
Dialogflow is a natural language understanding platform that makes it easy
to design and integrate a conversational user interface into your mobile app,
web application, device, bot, interactive voice response system, and so on.
Using Dialogflow, you can provide new and engaging ways for users to
interact with your product.

Dialogflow can analyze multiple types of input from your customers,

including text or audio inputs (like from a phone or voice recording). It can
also respond to your customers in a couple of ways, either through text or
with synthetic speech.

2.2 Signup For Dialogflow:

You need to have a google account to signup for Dialogflow.

a) Go to https://dialogflow.com/ and click on ‘Sign Up for Free’ button.

 b) Click ‘sign-in with Google’.

c) Select your google account and once you are redirected, click on ‘Go To
Console’ on the upper right corner of the screen.

2.3 Dialogflow Console:

Dialogflow provides a web user interface called the Dialogflow

Console (open console). You use this console to create, build, and test
agents. The uses are:

 Create Agents

 Create Intents

 Create entities

 Fulfillment to connect to other APIs.

  Integrate the bot with other platforms

 Analyze agent performance

 Test the agent in conversation simulator.

3. Building a chatbot using google Dialogflow

3.1 The problem statement:

To build a chatbot which can answer all the queries of a customer and
whenever a customer does an enquiry, it automatically send the customer
the course details. Also an email is sent to the support team to assist the
customer futher with their queries.

3.2 Agent:

A Dialogflow agent is a virtual agent that handles conversations with your

end-users. It is a natural language understanding module that understands
the nuances of human language. A Dialogflow agent is similar to a human
call center agent. You train them both to handle expected conversation
scenarios, and your training does not need to be overly explicit.

3.2.1 Creating an Agent:

a) Click on Create Agent from the left menu.

b) Provide the name of the agent and click on the SAVE button to create
the Agent.
 c) After saving, the agent is shown in the left hand side of your console.
You can click the gear icon (highilighted in orange) to edit the agent
settings.

3.3 Intent:

An intent categorizes an end-user's intention for one conversation turn. For

each agent, you define many intents, where your combined intents can
handle a complete conversation. When an end-user writes or says something,
referred to as an end-user expression, Dialogflow matches the end-user
expression to the best intent in your agent. Matching an intent is also known
as intent classification.

The following diagram shows the basic flow for intent matching and
responding to the end-user:
3.3.1 Creating an Intent:

a) Click the + add button next to Intents in the left sidebar menu.

Enter a name for your intent. Your intent name should
represent the end-user expressions it recognizes and click Save.

3.3.1.1 Training Phrases:

Training phrases are example phrases for what end-users might

type or say, referred to as end-user expressions. For each intent,
you create many training phrases. When an end-user
expression resembles one of these phrases, Dialogflow matches
the intent.

Adding the training Phrases:

a) Click the text field that shows "Add user expression".

b) Type your training phrases and press the Enter key after
each.
3.3.1.2 Extracting the Entities:

When an intent is matched at runtime, Dialogflow provides the

extracted values from the end-user expression as parameters.
Each parameter has a type, called the entity type, which
dictates exactly how the data is extracted. Unlike raw end-user
input, parameters are structured data that can easily be used to
perform some logic or generate responses.

Each intent parameter has a type, called the entity type, which

dictates exactly how data from an end-user expression is
extracted.

Dialogflow provides predefined system entities that can match

many common types of data. For example, there are system
entities for matching dates, times, colors, email addresses, and
so on. You can also create your own custom entities for
matching custom data.

3.3.1.3 Specifying custom Responses:

 Intents have a built-in response handler that can return
responses after the intent is matched. This feature only
supports static responses, though you can use parameter
references in these responses to make them somewhat dynamic.
This is helpful for recapping information provided by the end-
user. For example, your intent response could look like:
"Okay, I booked a room for you on $date".

3.4 Fulfillment:

By default, your agent responds to a matched intent with a static response. If

you're using one of the integration options, you can provide a more dynamic
response by using fulfillment. When you enable fulfillment for an intent,
Dialogflow responds to that intent by calling a service that you define. For
example, if an end-user wants to schedule a haircut on Friday, your service
can check your database and respond to the end-user with availability
information for Friday.

Each intent has a setting to enable fulfillment. If an intent requires some

action by your system or a dynamic response, you should enable fulfillment
for the intent. If an intent without fulfillment enabled is matched, Dialogflow
uses the static response you defined for the intent.

When an intent with fulfillment enabled is matched, Dialogflow sends a

request to your webhook service with information about the matched intent.
Your system can perform any required actions and respond to Dialogflow
with information for how to proceed. The following diagram shows the
processing flow for fulfillment.
a) The end-user types or speaks an expression.
b) Dialogflow matches the end-user expression to an intent and extracts
parameters.
c) Dialogflow sends a webhook request message to your webhook service. This
message contains information about the matched intent, the action, the
parameters, and the response defined for the intent.
d) Your service performs actions as needed, like database queries or external
API calls.
e) Your service sends a webhook response message to Dialogflow. This message
contains the response that should be sent to the end-user.
f) Dialogflow sends the response to the end-user.
g) The end-user sees or hears the response.

The entire code for fulfilment can be found at

https://github.com/viratsagar/iNeuron_academics_bot

3.4.1 Functionalities achived in fulfilment:

a) Sending an email to the customer with the syllabus and all the course
details based on the course selected.

b) Sending an email to the Support team to further contact the customer

for further clarification.

3.4.2 Webhook for fulfilment:

a) Select Fulfillment in the left sidebar menu.

b) Toggle the Webhook field to Enabled.
c) Provide the details for your webhook service in the form. If your
webhook doesn't require authentication, leave the authentication
fields blank.
d) Click Save at the bottom of the page.
3.4.3 Fulfilment Request and Response:

To see the request being sent to the webhook call, click the ‘Diagnostic
info’ button.

After clicking, you’ll see the following structure:

"responseId": "f1f8df5c-0fdf-43cc-aca9-a41bfd70acfa-ab1309b0",

"queryResult": {
 "queryText": "1234567890",

"parameters": {

"cust_name": "vss",

"cust_email": "",

"course_name": "DataScienceMasters",

"cust_contact": "1234567890"

},

"allRequiredParamsPresent": true,

"fulfillmentText": "We have sent the course syllabus and other

relevant details to you via email. An email has been sent to the
Support Team with your contact information, you'll be contacted
soon. Do you have further queries?",

"fulfillmentMessages": [

"text": {

"text": [

"We have sent the course syllabus and other relevant details to
you via email. An email has been sent to the Support Team with your
contact information, you'll be contacted soon. Do you have further
queries?"

],

"intent": {

"name": "projects/faq-goilum/agent/intents/64b7ea04-23a9-4a5a-
bcc5-08e971134d9c",

"displayName": "course_selection"

},

"intentDetectionConfidence": 1,
 "diagnosticInfo": {

"webhook_latency_ms": 4731

},

"languageCode": "en"

},

"webhookStatus": {

"message": "Webhook execution successful"

The Request above gives you the idea of how to parse the requests in
your webhook service.

We’ll discuss the webhook code in detail during the session.

3.5 Knowledge base:

Knowledge connectors complement defined intents. They

parse knowledge documents (for example, FAQs or articles) to find
automated responses. To configure them, you define one or
more knowledge bases, which are collections of knowledge documents.

For this project, we have created a knowledge document which

contains all the FAQs of the sudents and we have integrated the same
with the bot.

3.5.1 Creating a Knowledge base

a) To create a knowledge base, click on ‘Knowledge’ on the left hand

side of your console and then click ‘CREATE KNOWLEDGE
BASE’.

b) Give the name of your knowledge base and click ‘Save’.

c) Once created, you’ll see a message stating that no knowledge
document has been created yet. Click ‘create the first one’ to start
creating your document.

d) Select FAQ and text/CSV and the upload the FAQ CSV file
created and then click Create.

e) Scroll down to the Responses section and add responses as desired:

 When defining the first response for the intent,
use $Knowledge.Question[1] and $Knowledge.Answer[1
] where you want the question and answer to be
supplied.
 The index
for $Knowledge.Question and $Knowledge.Answer start
s at 1, so increase this index when adding more
responses.
 f) Click SAVE once you are done editing.

3.6 Test the Agent:

You can now test the agent from the ‘test it now’ section.

3.7 Integration:
 Dialogflow integrates with many popular conversation platforms like
Google Assistant, Slack, and Facebook Messenger. If you want to
build an agent for one of these platforms, you should use one of the
many integrations options. Direct end-user interactions are handled
for you, so you can focus on building your agent.

3.7.1 Integration with Telegram:

Dialogflow Telegram Integration allows you to easily create Telegram

bots with natural language understanding based on the Dialogflow
technology.

In order to set up the Telegram integration for your agent, you'll

need the following:

a)  Telegram account.

3.7.1.2 Creating a Bot in Telegram

a) Login to Telegram and go to https://telegram.me/botfather

b) Click the Start button in the web interface or type /start

c) Click on or type /newbot and enter a name
d) Enter a username for the bot, ending in "bot" (e.g.
garthsweatherbot)
e) Copy the generated access token
3.7.1.3 Setting Up Dialogflow

a) In Dialogflow, go to Integrations in the left hand menu

b) Click on the Telegram tile

c) Paste the Access Token into the related field

d) Click the Start button
Now you can search your bot in telegram and you can chat with it.