AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY

(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)

NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

AI CHAT BOT UNIT -1 NOTES

1. How, The Chatbots are useful for Business?

A) What is Chat bot?

An AI chatbot, or artificial intelligence chatbot, is a computer program or application that uses
artificial intelligence and natural language processing (NLP) to interact with users in a
conversational manner. The primary purpose of an AI chatbot is to simulate human-like
conversations and provide automated responses to user queries or requests.

AI chatbots can be designed to operate in various platforms, such as websites, messaging apps,
social media, or even integrated into voice-based systems. They use algorithms and machine
learning techniques to understand and interpret user input, and then generate appropriate responses
based on the context of the conversation.

These chatbots can be categorized into two main types:

1. Rule-based chatbots: These chatbots follow a predefined set of rules and patterns to respond to
user queries. They are typically less flexible and rely on a preprogrammed decision tree to
determine their responses.

2. Machine learning-based chatbots: These chatbots use artificial intelligence and machine learning
models to learn from data and user interactions. They can adapt and improve over time, providing
more dynamic and contextually relevant responses.

AI chatbots find applications in various industries, including customer support, e-commerce,

healthcare, finance, and more. They can help businesses enhance customer service, streamline
communication, and provide quick assistance to users at any time of day without the need for
human intervention.

It's important to note that the capabilities and performance of AI chatbots depend on the
sophistication of their underlying AI models and the quality and quantity of data used to train them.
As AI technology continues to advance, chatbots are becoming increasingly capable of
understanding and engaging in more natural and complex conversations with users.

Now We are Using the Chatbots for our Business with help of Technology along with AI and
Business Process Logic. At the same time the Research states that According to the Grand View
Research 2018 report. The Global market of Chat bot reach $1.25 billion by 2025, with 24.3%
average annual growth rate.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Do you know? The necessity of Chatbots market will grow across the financial service sectors in
customer – facing business because the chat bot will reduce the operation cost and give customer
satisfaction.

Some points to remember how the chatbots are helping to Business.

• This Chatbots Technology provides interface between exchanging messages between

chatbots and customers like carrier. The chat bots can be used for business through mobile
phone or web in real time.

• Here we should remember that the AI builds the Core brain for Chatbots, which is used to
understand the natural language decode form machine instructions. At the same time
making decisions during conversations.

• The most important part is business process design, which identifies the standard process
to access information, its means what information can be shared with whom, and
convenient ways to buy /sell/inquire about current products.

Likewise, chatbots gives more benefits in terms of reducing cost, increasing the customer's
satisfaction, produce revenue by sales. Literally, adds immense value to customers’ experience.

• 24X7 availability: Chatbots are available 24X7 through phones or the web. Customers can
interact and get clear queries with company services through the chatbots.
• Zero human touch experience: The customers can solve their queries and concerns by
interacting with humans in those issues. They can get the information with fingertips
against manual process.
• Simplicity: The Customer can get their information in simple steps and very concise as
per customer query.

Task: Go through the link below and add some points to the above question.
Chatbot Market Size Worth $27,297.2 Million By 2030 (grandviewresearch.com)

2. Explain the A Customer – Centric Approach in Financial Services?

A) The Customer – Centric Approach means it is an process of involvement of direct and indirect
innervations (involvement) through multiple channels.

For the past last two decades the customer behavior and interactions have changed to personalized
approaches. Due to the competition to give more service to customers and more depends on
technology these are main reasons for change the customer interventions.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Customer centricity has a life cycle. Now we have some indicative steps to reach the customer
centric approach.
1.Customer – Focused Leadership: it is the core element of customer centricity. If the leadership
aligns the strategy to become a customer centric organization, the whole outlook and
communication become customer – centric. The Amazon company has proved this method. The
now consider as benchmark for a customer – centric approach.
2.Understanding your customer: Understanding the customer and designing experiments to
validate the opinion from the next steps in a customer centric approach. We should understand the
customers’ all types of requirements we have to set up a successful connection with customer.
Then move to the next step.
3.Design the experience: Once all set up is completed, we should design the experience from
customer. And make sure every customer validates this response.
4.Empower frontline: Once all steps are done we need to analyze and set front line with strong set
up in terms of customer.
5.Metrics that matter: Metrics means measurement of tracking our progress. We should analyze
metrics of above our set up and designing process. As well the essential metrics.
6.Feedback: Every time take the feedback form customer because based the feedback we move
and improve ourselves.

Now we discuss some problem while using the services/products. In the financial services, majorly
the focused on the retail’s products and/ services. Here the interaction points and all touch points
are many to customer to focused. As well as the many bank and insurance companies dealing with
many customers via multiple channels.

Here we mentioned some critical trends among banking and insurance services while customers
are accessing.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

More natural interactions: The user experience is of the utmost importance. The customers are
looking for easier access to products, an appealing experience, and easy action in a few clicks.

More touchpoints and flexibility: The customer do not want fixed 9 am to 5 pm branch visits or
no access on weekends. Customers want to be able to access and buy products anytime and via
multiple channels. It may be a mobile app or a web app or phone banking, but they want more
flexibility in how they interact.

Responsive service: Customers expect that the bank/ insurer knows about them and is responsive
to their needs. They want individual attention and appreciate responsive customer service.

Clear product information: With so many players and products, customers want concise and
relevant information to be delivered to them. Additional details they can seek with follow-ups. The
customer does not want a pile of information or to get confused.

Great value from the products: The product features are numerous and many times the customer
is unaware of how to make the best use of them. Customers expect the bank/insurer to keep
reminding them to draw the best value out of a product and if possible, offer new products that
might be useful.

3) How, The Chatbots are important in Insurance industry?

A) In the past, we had to wait for someone to help us or put in a lot of effort to find information.
Nowadays, companies focus on putting customers first, which sets them apart from others.
Chatbots are computer programs that help interact with customers and promote the company's
image. They are useful in many industries, including insurance. Mobile phones and social media
have not only given us new ways to communicate but also made us feel more connected to
businesses. Companies are spending a lot of money on creating and managing a strong online
presence and using new tools to reach more customers.

In the past, the insurance industry was slow to change because it dealt with many different
risks and had a complicated way of operating. There was a lot of paperwork, background checks,
and approvals involved. But now, with the rise of digital technology and more competition, the
insurance industry is adapting to the needs of the always-online and always-connected digital world.

About 70% of the calls made to an insurance company's call center are questions that can be
answered without talking to a person. These questions include things like checking the status of a
claim, renewing a policy, or getting information about financial advisors.

According to Worldwide Call Centers, the cost of incoming calls can range from $0.35 to $0.45
per minute for low-cost international agencies, and from $0.75 to $0.90 per minute in the U.S. and
Canada. Internationally, the cost can be between $8 and $15 per call, while in the U.S. and Canada,
it can be between $22 and $28 per call. A big insurance company typically receives more than 10
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

million calls each year. If we assume the cost of a call is $5, and if a chatbot can handle half of these
questions, that could save the company approximately $25 million per year.

Some Important points to remember:

1. Chatbots can handle multiple conversations at once, unlike live agents who can only handle a
few.
2. Chatbots can automate repetitive tasks.
3. People usually have to wait around 3 minutes before speaking to a call center agent.
4. It takes people about 5 to 10 minutes to find information on a website.
5. Virtual agents like chatbots provide instant information.
6. Chatbot services are available 24/7 on different platforms such as mobile apps, Facebook
Messenger, Twitter, SMS, Skype, Alexa, and web chat.

Concise explanation:
Live agents can only handle a few conversations, but chatbots can handle many at the same time,
reducing the need for as many human workers. Chatbots can also do repetitive tasks automatically.
When people call a call center, they usually have to wait for about 3 minutes to talk to an agent.
Searching for information on a website takes people around 5 to 10 minutes. However, virtual agents
like chatbots provide instant information, making interactions faster and more efficient. Chatbot
services are available all the time on different platforms, so customers can get support using various
channels.

Task :

Add some points about call centers:

Go through below link:
2
www.worldwidecallcenters.com/call-center-pricing/

Figure shows the ways in which chatbots are transforming the insurance industry.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Instant

AI-Based
Instant

Experience

in

Buying
Process

Queries

Advisor

Automated Underwriting
Companies are using smart technology to accurately understand how risky someone is by looking
at the information available about them online. They use computer programs called chatbots to make buying
life insurance easier and give quick decisions.

Instant Quotations
Customers can get insurance eligibility and quotation details on the platform of
their choice instantly.

AI-Based Personalized Experience

Chatbots are computer programs that act like humans, and they use AI to understand
what people mean and what they need. This helps them give better customer service
and make people happier with their experience.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Simplification of the Insurance Buying Process

Many people don't like filling out complicated insurance forms because they can be
confusing. Chatbots can make things easier by asking simple questions in a friendly
way. They can then fill out some parts of the online form automatically using the
answers you provide. This makes the application process faster.
Registering a Claim
Since chatbots are virtual agents, they are available 24×7. Hence, they can help
customers with the claim process regardless of the time of the incident. Time-to-
settle-claim is an important metric that plays a critical role in improving the
efficiency of an insurance business, and chatbots are playing an essential role in
reducing the overall time to settle.
Finding an Advisor
Some companies use insurance agents or advisors. The natural language- based
interaction model of the chatbots makes them convenient for customers to quickly
inquire about insurance agents or financial advisors based on location or insurance type.
Answering General Queries
Around 30% of the questions people ask the call center are general questions about things
like the cash value of a policy, when the premium is due, interest rates, frequently asked
questions, information about the company and products, and issues with their account,
such as resetting a password or updating the beneficiary. A chatbot can answer these
questions right away and in a way that focuses on the customer.
Policy Status
The customer can check policy status and statuses of claims or other complaints or
requests by interacting with a chatbot anywhere anytime.

Instant Notifications

Chatbots can remind customers about the policy premium due date, next billing cycle,
and so forth.

New Policy or Plan Suggestions

Chatbots not only perform the role of service agents but also provide new marketing
opportunities. User interaction and social media behavior on digital platforms such
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Facebook can be tailored to suggest content, products, or services as per their needs.

4.Explain the Conversational Chatbot Landscape and its methods.

A) In today's world of the internet, whenever we need a service or information, we usually

go to a website. With smartphones becoming popular, we also have apps for the same purpose.
Almost every business now has at least one website and one mobile app. However, in the era
of artificial intelligence (AI), there is so much information on these platforms, and there aren't
enough employees to help all the customers who need assistance. Even if a company hires
more employees, it can be very expensive. That's where chatbots come in. These special
computer programs called chatbots help solve the problem of too much information by
providing affordable assistance to customers.

The organization is making changes to improve its digital systems, and they are
considering using chatbots. The main goal is to make things easier for customers and provide
faster service. This can lead to more people buying new products and services and lower costs
for the company. Because of advancements in technology and AI, chatbots are being used in
many different industries. Companies are creating chatbots to help customers at every stage
of their experience.
• Acquisition

• Engagement
• Servicing
• Feedback

Companies focus on two important things: getting new customers and keeping them
engaged. This helps the company make more money. Additionally, by providing good
customer service, they can save money. Customer feedback is also important because it helps
the company keep customers coming back.

Figure 1-7 shows an industry-wide adaption of chatbots and various use-cases.

AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Insurance companies are now able to process claims 24% faster, which is a big benefit.
Telecommunication companies have also reduced customer service calls by as much as 90%.
They do this by using self-service guides and automatic tips for 75% of the calls, and for
another 15%, they use artificial intelligence. This means that only 10% of the calls need a
costly human operator. By answering 80% of common questions in a service center,
companies can reduce customer service costs by 30%. For example, Autodesk saw a 99%
improvement in response time for basic queries. These benefits keep increasing as more
companies use these methods and make improvements.

Task:

Go through the below links and add some points:

3
www.avaamo.com/

4
www.mckinsey.com/industries/telecommunications/our-insights/a-future-

for-mobile-operators-the-keys-to-successful-reinvention
5
https://chatbotsmagazine.com/how-with-the-help-of-chatbots-customer-

service-costs-could-be-reduced-up-to-30-b9266a369945
6
www.ibm.com/blogs/watson/2017/10/how-chatbots-reduce-customer-service-costs-by-30-percent/

5)How to Identify the Sources of Data and explain Chatbot Conversations.

A) Chatbots are like another way to talk to customers. In the last chapter, we talked about how banks and
insurance companies are organized and how they interact with customers. Banks and insurers have many
different points where they connect with customers during their daily work, like selling a new policy or
dealing with customer complaints. All these interactions provide data that helps create an AI Assistant,
which is the chatbot. In this chapter, we will learn about the different types of chatbots and where the data
comes from to train them. We will also discuss the General Data Protection Regulation (GDPR) and how it
relates to personal data used by chatbots.

Chatbot Conversations: The chatbot tries to talk like a real person. When it interacts with a business, the
conversations can be about general things or about specific products or services. We can divide these
conversations into two types: general and specific. The type of conversation determines what the chatbot
or human assistant needs to know to talk to a customer.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

General Conversations: A general conversation is when a customer and assistant talk about various
things without sticking to one specific topic. It can go in any direction based on what the assistant knows.
For example, when a customer visits a bank to talk to the manager, the conversation can be about different
things like a sponsorship event, a loan account, or utility payment.
To handle such conversations, the chatbot needs to be built with different types of contexts and
suitable replies. These replies are not just based on simple rules; they require human-like intelligence and
information that the chatbot doesn't have access to.
Even though AI is improving a lot, creating chatbots that can have complete general conversations
like humans in real-life situations is still a challenging task, especially for industries. We are making
progress, but we are not there yet.

Specific Conversations: Specific conversations have clear and limited goals. These talks know exactly
what they want to achieve and have clear instructions to follow or refer to other sources if needed.
If there are any deviations from the set conversation, they are usually directed to specific outcomes.
Otherwise, they may be redirected to the appropriate place, or the conversation ends.

For example, when a customer goes to a refund desk in a store, the conversation is focused on
getting a refund. If the customer asks about discounts, they are directed to another counter because that's
not the purpose of the refund desk.

Specific conversations are more predictable and can be handled accurately. Chatbots designed for
specific tasks can provide information and focus on achieving a particular goal. Once the goal is achieved,
the conversation ends.

6) What type of training required for Training Chatbots for Conversations?

A) Chatbots need to learn how to talk to people. They are trained using rules and real
conversations. It's tough to create a chatbot that can have accurate general conversations like
humans. Examples of such chatbots are Alexa, Siri, and Google Home.

To make chatbots work well, we need a set of rules for how they should handle
conversations. For instance, if a chatbot asks for a customer's name, it should expect both the first
name and last name. If it doesn't get the last name, it should ask again to confirm.

To train a chatbot, we need a lot of training data. We gather this data from various sources
depending on what the chatbot will be used for. In the next sections, we'll talk about some datasets
used to train chatbots.

Self-Generated Data: To bring a chatbot to life, developers need to start with some data. They
create this data themselves to set the initial flow of the chatbot and work on it based on their
assumptions.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Developers often create various inputs and label them for training the basic chatbot
responses. However, this self-generated data is not enough for complete training. It helps prepare
the chatbot for a test release, where real users provide more data.

The data created by developers is used to test the chatbot's system integration. Once the
test version (beta) is live, internal users try it out, and more data is collected to improve the
chatbot's natural language understanding. This self-generated data is just the starting point and not
meant for public use.

7) Explain types of Customer Interactions?

A) Talking to customers is the best way to train chatbots. These conversations are great to copy
because of two main reasons:

1. We can collect common questions and focus on training chatbots for specific conversations.

2. The conversations also record real solutions given by experienced customer representatives in
the past.

Customers interact with businesses through different ways, and all these ways provide data to train
chatbots. Figure 2-1 shows six main types of customer interaction channels for any modern
business, which also apply to insurance and banks.

Figure 2-1. Customer interaction/service channels

 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Phone

Experienced call center representatives handle phone calls, especially when customers need quick help.
However, nowadays, this method is used as a last resort because it's expensive for companies. In the call
center, we can collect call transcripts, recordings, and the main problems customers have with their
solutions. This information helps our chatbot learn to recognize issues and give helpful answers.

Emails

Email conversations are usually long and explain what happened step by step, including what the customer
needs. They can be helpful for capturing complex problems that require more than simple data. We can
access customer emails in plain text, seeing the original email and all the back-and-forth responses that
create a complete conversation.

Chat

A lot of banks and other financial institutions use online chat with customer service representatives.
This way, they can help many customers at once and avoid missing calls at the call center.

This chat data is almost what a chatbot needs to learn how to talk like a human. We can get past
chatlogs in simple text files to use for training the chatbot.

Social Media

Social media became popular when companies could create business accounts on these platforms.
Interactions on social media are usually broad and not easy to track with specific customers or the
general public.

Customer Self-Service:

Some simple troubleshooting processes are made available for customers to use themselves. This
includes things like changing a PIN or finding answers to frequently asked questions. When these
self-service options work well, they help train the chatbot to assist people who can't or don't want
to use self-service. The data collected here is organized like a tree with different conversations
leading to specific solutions.

Mobile:

Mobile interactions refer to how customers use mobile apps and browse websites on their phones.
The data from these activities is recorded as logs of what customers do.

Customer Service Experts:

Customer service experts are people who know a lot about handling customer questions and
problems. Their insights are used to create default replies and backup plans for the chatbot. Their
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

experience is valuable for training and testing the chatbot. They are also involved in the
development process to check the chatbot's performance and accuracy.

Open Source Data:

Open source data is very helpful when making chatbots that can talk about many different topics. It provides
a lot of information for training chatbots to have natural conversations. Here are some examples of open
data sources you can use to train chatbots, but there are many more available based on what you need.

• Yahoo Language Data, created from Yahoo Answers

(www.cs.cmu.edu/~ark/QA-data/)
• WikiQA corpus, created from bing queries that redirect to wiki pages with
a solution (http:// research.microsoft.com/apps/mobile/download.
aspx?p=4495da01-db8c-4041-a7f6-7984a4f6a905)
• Ubuntu Dialogue corpus, created from Ubuntu technical support
(www.kaggle.com/rtatman/ubuntu- dialogue-corpus)
• Twitter data on Kaggle, created from customer support at Twitter
(www.kaggle.com/thoughtvector/customer- support-on-twitter)

Crowdsourcing:

The most important training data for a chatbot comes from real people talking to it in real time. This helps
train the chatbot and shows developers where it may have trouble.

In the best cases, chatbots are tested by real customers and employees before the final version is released.
The data collected during this testing helps improve the chatbot's understanding.

Experts in customer service also use crowdsourced information to improve how the chatbot responds to
different conversations.

If you're making a chatbot in a regional language, you'll need to use crowdsourcing to gather training data.
Some companies can connect you with people who will interact with your chatbot to help build its training.

7) How to Handle and what is need for Personal Data in Chatbots ?

A) Personal Data in Chatbots: When chatbots have human-like conversations, people may
unknowingly share personal information with them. This data can be at risk of unauthorized access
and violate privacy laws.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Customers can reveal personal data on purpose, like giving their account number and PIN for
checking their balance. Sometimes, they may accidentally share personal data when discussing
certain topics.

During training and deployment, personal data is collected, which can be vulnerable to hacking
and legal issues. However, this data is essential for creating customer-focused chatbots that can
access and provide real-time information securely.

Using personal data helps develop features like authentication, compliance with company policies,
and systems to retrieve customer information. This is crucial for building specific conversation
chatbots for banks and insurance companies.

Due to data privacy regulations like GDPR, some well-known chatbot services cannot be used
directly. So, companies need to develop their own chatbot frameworks while being mindful of data
regulations. It's important to understand these regulations to build chatbots that respect customers'
privacy and comply with the law.

8) What are the considerations regarding personal data when using chatbots, and how does GDPR come
into play?

Introduction to the General Data Protection Regulation (GDPR):

The GDPR is a new law that replaced an old one from 1995 about protecting data. The old law was different
because it only gave guidelines to countries, and each country had to make its own laws based on those
guidelines. But the GDPR is a direct law that applies to all countries in the EU at the same time. It is meant
to protect the personal information of people who live in the European Union. The law also covers how
personal data can be sent to countries outside the EU. With the GDPR, people have more control over their
personal information. They can decide if they want to share it or keep it private.

The GDPR became a law on May 25, 2018, and all EU countries agreed to follow it. If organizations don't
obey the law, they can get big fines. The fines can be up to 4% of their yearly income or 20 million euros,
whichever is more.

Data Protected Under the GDPR:

The GDPR is a law that includes many types of data collected by companies. The protected data includes
personal information like name, date of birth, phone number, address, and more. It also covers web data
like location and IP address, health and genetic information, and other sensitive details. Even data that
chatbots use in conversations is included in this law.

Data Protection Stakeholders:

 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

According to the rules, any company that collects and handles personal information of people living in the
EU must follow the GDPR, even if the company is not in the EU. This means that most global businesses
need to follow the GDPR rules.

The GDPR defines three important roles:

1. Data subject: This is a person whose information is being collected and processed.

2. Data controller: This is an individual or company that decides why and how personal data is collected
and processed.

3. Data processor: This is an individual or company that processes the personal data on behalf of the data
controller.

When we create chatbots, we need to consider these roles to protect our customers' rights. For example,
when a customer talks to a chatbot, they are the data subject, and the bank or company becomes the data
controller. The system that stores the data also acts as the data controller. If the chatbot uses Dialogflow to
process data, then Dialogflow becomes the data processor.

The details of the law can be read from the source here: https://eur-
lex.europa.eu/eli/reg/2016/679/oj.

Customer Rights Under the GDPR:

The chatbot developer team and leaders need to know and follow the rights given to
customers in the GDPR. The chatbot should work in a way that respects those rights.
The rights under GDPR are stated below for your reference:
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Important Questions for Unit -1:

1. Explain the benefits that businesses can derive from using chatbots. How can chatbots enhance customer
interactions and overall business operations?

2. In the context of financial services, discuss the importance of adopting a customer-centric approach when
implementing chatbots. Provide examples of how chatbots can improve customer experiences in this
industry.

3. Analyze the role of chatbots in the insurance industry. How can chatbots streamline insurance-related
processes and offer better support to customers?

4. Describe the landscape of conversational chatbots. Differentiate between rule-based chatbots and
machine learning-based chatbots. Discuss the advantages and limitations of each approach.

5. How is data sourced in chatbot conversations? Explain the methods used to gather and analyze data to
improve chatbot performance and user experience.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

6. Discuss the process of training chatbots for conversations. What are the key steps involved in training a
chatbot to engage effectively with users?

7. Personal data protection is crucial in chatbot interactions. Identify the types of personal data that chatbots
may handle during conversations. How can businesses ensure the privacy and security of such data?

8. Provide an introduction to the General Data Protection Regulation (GDPR). What is its significance in
the context of chatbots and customer data? How does GDPR impact chatbot development and usage?

9. Compare and contrast the benefits and challenges of implementing chatbots in different industries such
as finance and insurance. How can businesses overcome potential obstacles and maximize the advantages
of using chatbots?

10. As a chatbot developer, how would you ensure GDPR compliance in the design and functioning of a
chatbot? Outline the steps you would take to safeguard user data and uphold their privacy rights.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

UNIT 2
1. Explain below Chatbot Development Essentials:
i. Customer Service-Centric Chatbots.
ii. Business Context
iii. Policy Compliance
iv. Security, Authentication, and Authorization
v. Accuracy of User Input Translation to Systems
(Or)
What are the critical considerations in chatbot development to achieve the intended use case from a
chatbot?
(Or)
What are the critical considerations in developing chatbots to achieve their intended use case?

A) Chatbots are designed to talk with people in a way that feels like talking to another human. This makes
them better than traditional apps with menus. In earlier chapters, we covered different types of chatbots
and the rules that should be followed when making a chatbot for your own company.
Customer Service-Centric Chatbots Context:
Customer service chatbots are designed to help customers by exchanging information and accessing data.
To do this, the chatbot must be accurate and follow strict security and privacy rules.
There are three important factors to consider when developing a chatbot for a specific purpose:
Business Context:
This means understanding the unique needs of the business the chatbot serves. In regular
conversations, people can understand things that are not explicitly said. In the past, live chat systems were
used to reduce costs, but they sometimes led to poor customer experiences because the context had to be
repeated in every conversation. Modern chatbots need to recognize and maintain the context to have
smoother conversations. The context is like the unsaid information that a successful chatbot should
remember.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

For example, let's look at a conversation to see why context is essential

Conversation context: User chatting about the food order (2345) he canceled last
night.

Case 1: No Context Maintained

User: Why refund not processed for my food order?
Chatbot: Please provide your OrderID
User: 2345
Chatbot: Your refund is processed and will be updated in your account in 2 days.

Case 2: Context of Chat Maintained

User: Why refund not processed for my food order?

Chatbot: Your refund is processed and will be updated in your account in 2 days.

In Case 1, the chatbot doesn't remember what was said before, so in every conversation, you have to give it all the
information again. In Case 2, the chatbot remembers what was said earlier and tries to understand what you mean.

For example, if you're making a chatbot for a bank, it needs to understand special words and phrases used in banking
like "time deposit account" or "roll over." This is important because users don't want to keep explaining the same things to the
chatbot.

To make the chatbot understand all this, it needs training. It learns the specific business language and terms used in
the bank through Natural Language Processing (NLP). This training helps the chatbot to be better at conversations with
customers.

Note:Every conversation may have a global context and a dialogue- specific context. For example, talking about
life insurance is the global context, and within that, talking about a particular policy is the conversation
context.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Policy Compliance:

When using a chatbot, people can make different kinds of requests. But it's important to know the
right way to handle those requests. Policies and government rules decide what's allowed and not allowed,
and how to get the information needed.

For example, let's say someone wants to update their home address for insurance.

User: Update my home address to XYZ, sector-8, India

Chatbot: Address updated to XYZ, sector-8, India

Technically, it seems easy to just tell the chatbot the new address, and it confirms the update. But
it's not that simple. There are questions like: Is it allowed to change the address this way? What are the
proper steps at a real bank branch? What rules should be followed?

A policy guide is needed to help the chatbot follow the right process. It has to consider both business
rules and government regulations for any request to be done correctly. For example, changing the address
might need more than just a simple confirmation. It could require showing address proof, giving a reason,
waiting for a few days, or following other steps.

Customer service chatbots need to be carefully taught these rules to handle requests correctly. The
rules can be different for each business or purpose, and they must be strictly followed in the chatbot's logic
as “AND”.

Security, Authentication, and Authorization Simplified:

Chatbots can either access private information or provide public information in a safe way. To keep
conversations secure, there are two important things to consider: authentication and authorization.

Authentication ensures that the chatbot knows who you are and you're not talking to someone else
pretending to be the chatbot. Authorization makes sure the chatbot can only access the information it's
allowed To make sure data exchanged between users and the chatbot is safe, a secure communication
channel is used. This means the information is encrypted and sent through a very secure medium.

We can enforce security policies by using the HTTPS protocol, having a firewall, and following
industry best practices. The network technology and technical architecture take care of the security layer to
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

keep all conversations safe. In many cases, we can use existing security features from the business
framework, so we don't need to design them separately.

As seen in Figure 3-1, the user first needs to establish the identity via the authentication process.
The authentication process generally involves providing a username and password, and in multi-factor
authentication, additional information like an OTP or PIN. The successful authentication establishes the
user identity as per system records. The next process is to establish the permissions the user has in the
system. Authorization is also called role management in business terminology, and hence the business needs
to assign roles/permissions for each authenticated user before deployment of the chatbot.

Figure 3-1. Reference accountability framework

The chatbot needs to make sure users are allowed to use its features, and if they are, it checks which
features each user can use. This is especially important when the chatbot connects to the main system to get
information like HRM or CRM.

To do this, the chatbot uses authentication. It checks if the user is genuine and allowed to use the chatbot.
Sometimes, a system like Active Directory helps with this. Users might have to use multiple ways to log in
or use a special PIN to get access.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Authorization is another important part. It lets the company control who can access what resources. This
is essential because many users may want to use the same information, but we must protect the data and
only show what each user is allowed to see, based on their role and the rules in place.

Figure 3-2. Authentication and authorization service

Accuracy of User Input Translation to Systems:

The chatbot has to convert what users say in regular language into instructions that machines can
understand to get the right information. This translation needs to be very accurate before giving the answer
to the user. It's one of the biggest challenges for chatbots using NLP, and researchers are working hard to
improve it.

Let’s explain this with an example:

In this example, Input 1 and Input 2 need to get the same result from the system. The chatbot logic
needs to resolve for the two inputs before it can fetch the correct information for the user. Here are the
challenges and requirements of making sure the query gets accurate data for the system to respond. If the
chatbot logic is unable to create the right question, the results will not be correct and may even cause
unauthorized information to be shared with users unintentionally.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

2. Compare and contrast the rules-based approach and AI-based approach in chatbot development.

(Or)
Explain the Chatbot Development Approaches

i. Rules-Based Approach
ii. AI-Based Approach
(Or)

Compare and contrast the rules-based approach and AI-based approach in chatbot
development. Which approach is more suitable for specific use cases?

A) Chatbot Development Approaches:

Building a Chatbot: Finding the Right Balance When creating a chatbot, we need to decide how it will
work. We have to find the best way to make it understand natural language and give accurate answers. This
can be tricky because NLP has difficulty understanding regular conversations and turning them into
machine actions. So, we have to make sure to find a good balance in how the chatbot works.

As shown in Figure 3-3, there is a general classification of chatbots based on their abilities and the
extent of AI built within them. The critical axis over which we classify chatbots is conversation scope and
machine response.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

The focus of this book and our discussion is on closed domain conversations, where we can have
either a rule-based approach or create smart machines using an AI-based approach. In this section, we
describe the two popular types of development approaches.

Rules-Based Approach :

The rules-based approach, also known as the menu-driven approach, is like an improved version of
self-help portals. The main difference is how you find solutions. In a self-help portal, you have to click
through different options manually, but in menu-based chatbots, you can use regular language to navigate
and then perform actions with menus.

See Figure 3-4.

These kinds of chatbots are prevalent and usually high in usage across industry use
cases that are integrated with CRM and other data systems.

Figure 3-4. Menu-driven chatbot interface

 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

As shown in Figure 3-4, a chatbot attempts to understand the user question and then presents a menu to
choose the next action. The list makes sure the backend knows what exact operation needs to be done to
fulfill the request.

Advantages and Disadvantages of the Menu-Based Approach :

Advantages:

1. Responses are accurate and reliable.

2. It's simple and easy to understand and use.

3. New menu items can be added without major changes.

Disadvantages:

1. Limited to pre-defined options.

2. Requires two steps to fulfill a request.

3. Limited in handling natural conversations.

Even though it has some limitations, the menu-based approach is successful when accuracy is a top priority,
and a natural conversation experience is not essential.

AI-Based Approach :

The AI-based approach uses a smart NLP engine to understand natural language and find
information using machine learning and system integration. At first, the chatbot may not be very accurate,
but it gets better with time. The main difference from the menu-based approach is the NLP engine. It's like
a brain that understands what users say and decides what to do next based on that.

As shown in Figure 3-5, the critical role of the NLP engine is to extract information from the natural
language input. The accuracy of the information extraction is critical because it will decide the outcome of
the conversation and be persisted in the system. The NLP engine needs to extract the information required
to instruct the system to act. In the menu- driven approach, the user must engage with menus to select exact
details before the system can act.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Figure 3-5. The NLP engine extracts exact information based on ML techniques

Advantages and Disadvantages of the AI-Based Approach :

Advantages:

1. Smart conversations happen without many steps.

2. The NLP engine can handle new situations and lots of text.

3. The chatbot can learn and create custom responses.

Disadvantages:

1. Training and maintaining the NLP engine is complex.

2. Sometimes, responses may not be completely accurate.

3. A lot of data is needed for the chatbot to work well with the NLP engine.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

3.Explain the Conversational Flow.

(Or)

Explain the concept of conversational flow and its significance in designing user-friendly
chatbots.

A) Conversational Flow:
Chatbots for closed-domain applications are designed for specific purposes and tasks. To make sure
the chatbot can handle all kinds of user conversations, we need to plan and define its scope and different
paths it can take. This planning is crucial to follow the rules and provide access to the needed
information.
The conversation flow is like a map that shows all the possible events, choices, and results during
a conversation. This is useful when the chatbot needs to remember the context and give responses that
involve multiple steps.

An example of a conversation flow is shown in Figure 3-6.

 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

The flow starts with a welcome message and either provides a menu (if it’s a rule-based
chatbot) or the user asks a sentence (entirely AI-driven chatbot). Once the chatbot NLP logic
identifies which functionality .

the user requires, there is a decision point to take the user to that conversation path. If the
user wants to check the status of an order, his next conversational decision point is to enter the
order number. Once the chatbot receives a valid order id, the backend calls for action to retrieve
information for that order and take it back to the user. This flow maintains the context as well, so
if he needs to track another order, he need not start from the root but can just enter another order
id and the chatbot will know to track the order status.

In more advanced chatbots, you can pass multi-intents in one line, but technically the
chatbot will process the request in the same flow.

So “Track my order number 465” is a single input from the user, which should fetch the
same result. The multi-intent chatbots are difficult to build, and chances of error are high.

Creating a chatbot flow is essential since this defines the scope of features and improves
the accuracy of the chatbot for the user. It is imperative to make it clear to the user what the chatbot
is meant to do for them and maybe define the features beforehand. The exception cases can always
be transferred to default responses or a human executive.

4. Define and explain the key terms used in chatbots such as utterance, intent, entity, channel, and human
takeover.

Key Terms in Chatbots

The development of chatbots has become a full-bodied development process, which means it is
essential for you to understand the terminology of chatbots before you attempt to develop one. The key
terms used in chatbot development also have multiple variants as professed by leading chatbot platform
providers like Amazon, Google, etc.

In this section, we will discuss some key terms frequently used in chatbot development. In further
chapters, we will use these concepts and terminologies to show how to develop the chatbot from
scratch.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Utterance

Utterance refers to anything that user inputs to the chatbot. The full end to-end input makes an
utterance, such as “Get me status of my order id 345,” “What is the temperature today?,” “Hello,”
“Good Morning,” etc.
Utterances are used to develop a classifier for intents in development. The chatbot stores as many
possible utterances in the database, which are the questions asked by the user, and clusters them
in different intents, which represent what the user wanted to say.
For developing a chatbot for insurance, we will need to capture actual questions asked by users
from different channels, such as chats, emails, office visits, customer value center, etc. We will
use all this historical data to train the chatbot to know the actual requirement of the user and
which conversational flow to use.

Intent

The intent is the intention of the user identified from the utterance captured by the chatbot. Identifying
intent is the essential function of a chatbot. In menu-driven chatbots, the menus help the user to pinpoint
the intent, while in AI-based chatbots, identifying intent is a task done by the NLP engine.

The successful matching of intent decides the flow of conversation and delivers the correct response to the
user. In domain-specific chatbots, the intents could be different from general intents, and hence domain-
specific training is required.

For example, for the utterance of “Show me the stock price of Apple,” the intent is to find the stock price.
Let’s call this intent as showStockPrice. showStockPrice is the main intent of the user, while the term
“Apple” is the entity, also called a slot.

Entity
Entity gives meaning to the intent by providing additional value to the utterance. An entity can be defined
as subordinate to the intent, which tells us the intent is related to which subclass. In this example, “Apple”
is the entity of @company_name for the intent #showStockPrice.

Entities, or slots, when maintained in sessions, help retain the context of the conversation. In this example,
after this first utterance we replied with the price. Just after that, the next utterance could be “and for
Microsoft.” In this case, the chatbot has already captured the intent as show Stock Price, so the slot changes
to Microsoft and the chatbot can fetch a stock price of Microsoft.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Channel

The channel is the medium used by the chatbot to connect with the user and fulfill their request. Nowadays,
all social media messengers allow chatbots for conversations (e.g., Facebook Messenger, Slack, Skype,
etc.).

However, for an application like our 24x7 Insurance Agent, we want to have our developed channel remain
compliant with privacy laws and

provide an added layer of security when we access the private information of users.

Human Takeover

The human takeover is a term used to denote human fallback during a conversation. Modern chatbots come
with the feature of fallback to

human assistance when the chatbot fails to understand intents and extract entities, or the confidence is NLP
output is low.

Human takeovers can be of two types:

A human takeover by choice: At any moment the user can choose to talk to a human, maybe because they
are more comfortable with humans or the chatbot isn’t solving their issues.

A human takeover by confidence: A confidence filter may decide if we can fulfill the request with high
confidence; if not, the request is automatically transferred to a human without a choice for the user. This
provides a seamless experience for the user.

5. Illustrate a use case for a 24x7 insurance agent chatbot and its benefits.

A. Use Case: 24x7 Insurance Agent

The chatbot that will be discussed throughout the book is based on the use case of an insurance
agent. AI-driven chatbots will be discussed using the multiple aspects of security, natural language
ML techniques, deployment, and business purposes. Each chapter will discuss some component of
developing this chatbot.

Now you know the essential conditions for planning the 24x7 Insurance Agent, including setting the
business context, type of development policies, and other considerations. In this section, we will
define the aspects for 24x7 Insurance Agent, due to scope all features will not be explicitly
implemented.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Business context: The 24x7 Insurance Agent will be able to have necessary conversations in the
insurance domain. People may ask about their policies, premiums, etc.

Policy compliance: The policies can be taken from the standard procedure followed at the
customer care center.

Security, authentication, and authorization: We can create a PIN-based authorization, while an

IDS-based authentication can also be used if the users already have a product.

Accuracy of user input translation to systems:

To ensure this, we can create a confidence filter-based human takeover mechanism.

The AI-based approach or menu-based approach: Both should be fine as per need and balance
between flexibility and accuracy.

Conversational flows: The flows must be created by sitting with the business and exploring its
policies.

The NLP training will require data of old conversations, intent list, most frequent entities, etc.

These decisions will help the developer to select the right structure and architecture for the chatbot
solution.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

UNIT III
Building a Chatbot Solution

1. Evaluate the advantages and disadvantages of chatbots compared to traditional mobile apps.

(Or)

1. Compare chatbots with traditional mobile applications (apps) and discuss the growth of
messenger applications in recent years.

(Or)

1. What are the business considerations for building a chatbot solution?

1. What are the differences between chatbots and apps?

A. Business Considerations: Any business will ask the question, “What business value does a
chatbot add?” This question is to be answered with objectivity and time targets. Technological
advancements may allow us to implement advanced chatbots and other solutions, but how they
add value to the business is a very subjective call. The business needs to evaluate all factors to
ascertain how a chatbot is good for their business.

Chatbots vs. Apps: From technological point of view, the business must tackle an important
question, specifically relevant to closed domain chatbots: whether to go for an app or a chatbot. In
terms of functional features, both can provide the same information for a given feature set. The
key differential happens to be chatbots being conversational in nature, while apps are self-service
applications.

The key considerations of chatbots vs. apps as mentioned by 2018 State of Chatbot
Report (www.drift.com/wp-content/uploads/2018/01/2018- state-of-chatbots-
report.pdf) are

• Chatbots are preferred to get quick answers for questions

and 24-hour access.
• Apps are preferred for ease of use and convenience.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

A survey in the report lists some factors that a business must check with their
current needs:
• Quick answers to simple questions
• Getting 24-hour service
• Convenience
• Quick answers to complex questions
• Ease of communication
• Ability to easily register a complaint
• Getting detailed/expert answers
• A good customer experience
• Friendliness and approachability
• Having a complaint resolved quickly

Growth of Messenger Applications

Another factor driving the need for chatbots is the increasing usage of messenger
applications and a stable growth rate of having a mobile- focused approach by companies.
The customer is now connected 24 hours to the Internet through their mobile handsets and
wants to access services via easy-to-use interfaces.
In early 2011, messenger applications started coming up and had good adoption rates as
mobile devices, internet connectivity, and cloud computing also picked up at same time
(www.businessinsider.com/ the-messaging-app-report-2015-11). Sometime in early 2015 we
got to the point where messenger applications, specifically WhatsApp and Facebook
Messenger, were at meteoric adoption rates and equalled the activity on social networks. As
the trend suggests, people are more active on messenger applications compared to social media
networks.
This trend points out the shift in consumer behaviors where chat is a preferred mode of
communication. And this implies that businesses need to enable this channel of
communication with customers as well using either chatbots or human chats.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Direct Contact vs. Chat

The increasing use of messenger apps has shifted the way customers want to interact with
businesses. In the early days of messenger apps, the studies showed an increasing preference for
contact via chats.

The survey by BI Intelligence as summarized by Chatbots Magazine

(https://chatbotsmagazine.com/chatbot-report-2018-global-trends- and-analysis-4d8bbe4d924b)
shows that as early as 2016, the mature market customers adopted chats very quickly. The
consumers believed they could solve their issues faster over chat than calling a customer care
representative. They also felt more comfortable with chats because they could keep them as a
record for follow-ups, unlike phone conversations.

Business Benefits of Chatbots

Considering essential business aspects allows a company to decide whether it wants to go ahead and build
a chatbot or improve current apps/ channels. If the company decides to go ahead with a chatbot solution, it
needs to understand the key value creation by chatbots. Chatbots are now accepted by users with different
percentages in different industries, with different adoption rates.

In the aforementioned Chatbots Magazine questionnaire, which asked how comfortable you are with being
assisted by an AI-based chatbot for business communication, the response showed that people are most
ready for such conversations for online retail, generic healthcare queries, and telecommunications.

Further, in the 2018 State of Chatbot Report, the top reasons for not preferring chatbots are the need for
assistance from a real person, less awareness of chatbots, and possible blocks due to lack of accessibility of
channels (i.e., not having a Facebook account or access to a smartphone).

The studies indicate that there is a value in adding this channel if the benefits for the company are high and
your customer is comfortable being assisted by AI-based chatbots. The two topmost value creations from
chatbots are discussed in following sections.

Customer Experience

Customer experience is the second most impactful factor for introducing chatbots to a
business. The customer experience brings a multitude of values to the business, not just
limited to direct sales or savings. The benefits of good customer experience include
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

• High brand value and recall

• High lifetime values (improved engagement)

• Brand differentiator from competitors

There are many other derived factors due to a happy customer experience. A loyal
customer base is a recipe for long-term success.
The Chatbots Magazine summary points out features that contribute to a good and
unique customer experience derived from chatbots. The key points are listed below for
reference:

• 24-hour service

• Getting an instant response

• Answers to simple questions

• Easy communication

• Complaints resolved quickly

• A good customer experience

• Detailed/expert answers

• Answers to complex questions

• Friendliness and approachability
The variety of new features attracts customers and creates unique value for
companies.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

2. Discuss the business benefits of chatbots and the metrics used to measure their success, such as
customer satisfaction index, completion rate, and bounce rate.

(Or)

2. What are the business benefits of using chatbots, and how can success metrics like customer
satisfaction index, completion rate, and bounce rate be measured for chatbot performance?

Success Metrics
Success metrics are important to define at the start of any chatbot development. The
metrics work as a compass to direct the solution and the intended benefits of the
chatbot. While there are success metrics related to the accuracy of the NLP engine,
the intent classifiers, and other technical aspects, in this section we will only talk
about success metrics from a business perspective.
The success metrics need to be manageable and measurable with a simple
explanation to the business. We will discuss a few metrics that can be used to track
and manage the success of chatbots. The metrics focus on success when you compare
a chatbot interaction with a human interaction.

Customer Satisfaction Index

The Customer Satisfaction Index (CSI) measures a customer service representative’s
quality of interaction by following up with the customer with a small survey and
capturing their experience of the interaction. CSI is one of the most impactful metrics
to monitor because it provides not only the satisfaction scale but also the areas of
improvement.

Completion Rate
The Completion Rate (CR) is defined as the proportion of interactions with a chatbot
that ended as the solution resolved for the customer. This metric tells us how many times
the chatbot can complete a conversation and deliver the required responses to the user.
A higher completion rate indicates a more efficient chatbot service.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Bounce Rate
Bounce rate (BR) can be defined as how many users move away from chat after typing
in one or two inputs to the chatbot. A high bounce rate means the chatbot is not
successful in engaging the user and this must reflect in some of the customer feedback.
Along with bounce rate, we also measure the reuse rate (RR), which refers to
how many customers come back and use the chatbot again. BR is a perfect metric to
identify those people who more tend to use chatbots and target similar customer
segments.

Managing Risks in Chatbots Service

New technology channels do bring risks. Customers and companies need to

understand the risks involved with using chatbots for any transaction or information
exchanged through chatbots. The risk is to be understood, communicated, and
mitigated before general customers are allowed to use the chatbot services.

Third-Party Channels

Banks and other financial institutions are very much aware about opening a new
channel for users to access financial services. While it adds convenience to the
customers, it brings some risks as well. Technology is growing way faster than the risk
frameworks we have. By the time a general user is able to figure out the risk with usage
or best practices for using chatbots, they may already face a security breach.
Top security risks arise from the communication channels for chatbots because they
are external to the bank’s security control. For example, a customer interacting through
Facebook Messenger is interacting with the bank systems using the Facebook platform,
which may have vulnerabilities and is not designed for banking operations, just generic
chats among people.
In a 2018 survey undertaken by Synopsys, 36% of respondents indicated that
customer-facing web applications remain the top security risk to businesses in the
Asia-Pacific. September 2018’s admission by Facebook that a security breach had
affected more than 50 million accounts came as a timely reminder that even tech giants
aren’t spared. (Source: finews.asia).
These cases require financial institutions to limit functionalities through
public channels for chatbot messengers. Developing an end-to-end chatbot
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

experience can reduce this risk as well, but adoption remains a challenge.

Impersonation
Another very prominent risk arises from impersonation. Impersonation can result in
similar looking chatbots, or humans having conversations using fake windows,
hacking social media, and other sources of impersonation. The banks already face a
lot of fraud due to criminal ingenuity from fraudsters and spend millions in education
for phishing, vishing, and other impersonation attacks.
Two-factor authentications are one possible way to reduce the impersonation
attacks by having two-step verification from two different sources. In most cases,
hackers are not able to crack both factors of authentication and are less likely to be
successful in fraudulent transactions.

Personal Information

Personal information revealed through the chatbot channels is a challenge for banks
to manage. It is challenging to control users who may accidentally enter their
personal information to get access to a service.

As the chatbots are driven by natural language, the chances of revealing personal details are
high.
The chatbots need to make sure that they use as little information as possible. It’s better to
create a PIN with adequate access control so that the user never needs to disclose personal
information; they can just use the PIN. Educating the users is an essential step to make sure
the user is aware and alert for any fraud.
Confirmation Check

Confirmation is the most impactful and sometimes last resort to make sure transactions done
via chatbots are legitimate. Fraudulent or mistaken transactions are possible using chatbots.
As new technology comes into user service, it takes time for the users to understand the right
use of the
service, and in this process, they might do some illegal transactions as well.
For any transaction that seems to be an anomaly or unexpected, it’s always good to call
up the customer and ask for confirmation before processing it. This check saves the user and
bank from fraud.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

3. Describe the generic solution architecture for private chatbots and explain its key components.

(Or)

3. Outline a generic solution architecture for private chatbots.

Generic Solution Architecture for Private Chatbots

Solution architecture is a practice of defining and describing an architecture of a system

delivered in context of a specific solution and as such it may encompass a description
of an entire system or only its specific parts. Definition of a solution architecture is
typically led by a solution architect.

In this section, we present a reference solution architecture you can apply, with
minor modification, to the ideas presented so far (Figure 4-1).

Figure 4-1. Solution architecture of the 24x7 Insurance Agent

 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Workflow Description
Here is the workflow:
1) Conversation interface: We will develop the interface from
scratch; it does not depend upon third-party interfaces (e.g.,
Skype, Telegram, etc.). This will help us create customized
interfaces for the organization’s need and extend other features as
required. The interface can be an entirely new mobile app for the
iOS, Android, and Windows platforms.
2) MFA and Active Directory: The authentication system will be
built at the back end to authenticate devices (MS Intune), users
(Active Directory), and applications (by PIN). We will make it
on a node environment to allow integration to other identity
management services as well. In the natural form, we will only
have PIN verification to access the application.
3) NLP engine: The NLP engine will be built to accept text inputs
from a queue and extract intent and apply context to the
incoming query. Once the question is broken into the required
components, it’s sent to the bot logic.
4) Bot logic: This is the core handler of the incoming request. It will
have two core inputs before processing the request. The bot logic
does not call information services until it has satisfied the process
set as per the below two methods. If there isn’t enough
information for the bot logic to reply, it’ll ask the user for more
information.
a) Policy interaction flows: These flows are bank expert- designed
workflows for the incoming request (e.g., if someone asks for an update
of address, what are the essential steps for a reply?) The steps will make
sure the user complies with the steps to get an answer. This ensures that
all policies, statutory or internal, are followed by the bot, just like an
informal HR. Also, policies and FAQ can be defined here.
b) Machine learning: The request that requires a machine learning
algorithms to improvise the output is requested from here (e.g., can I
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

request a statement from March 15 to March 18?) This needs a machine

to apply appropriate logic to extract dates, an employee id, and an
existing account balance to create the right query to the information
system. Further, sophisticated features like the mood of the employee, the
urgency of the request, and the sentiment of the question will be built
here.
5) Information service: This is the place where
real-time information is fetched for the employee request. This
service handles all appeals and prepares responses AND can also
send a request for actions as well. The information service will talk
to three core data services:
a) Data lake/databases: If some data needs to be fetched from some
database or data lake of corporate.
b) Third-party APIs/ODBCs: Interacts with HR systems via
ODBC/APIs or some other method that exposes itself with REST APIs.
c) Human HR: If there’s a low score of confidence in the reply, it will transfer
the request to HR for a response through the chat interface.
6) Actions: The information service will also route the requests
asking for some work to the actions item management queue
(e.g., a request for leave for tomorrow). This request requires
an update in PeopleSoft or the HR system. All such change
requests will be routed to an HR approval (can be
direct as well); once approved, either by real HR or by a policy-
based rule, it will be sent to an action queue. HR can support
broadcast here, and they can be delivered by an actions workflow.
7) Updates in real systems: For authorized and approved requests,
we update them directly in the HR system and trigger a
notification to the chatbot user and also trigger emails and other
built-in process flows of the banking system.
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

Key Features
Below are the key features:
• Built for you: We will make the bot for specific needs, not fitting
those needs to existing bot frameworks.
• Data privacy by design: The bot is developed with data privacy by
design. It will be fully compliant with local laws and internal laws.
• Developed with a microservices architecture:
The entire application will be based on principles of microservices
and hence will allow future-proof design and also advanced
application development on top of the framework.
• Options for deployment: We can choose which components
we want to deploy on-premise or on the public cloud. Based
on needs, we can create a deployment plan.
• Extensible: We can integrate as many APIs or AI/ Ml features as
deemed fit for use. All the new future
changes can be consumed as APIs in the framework.

Technology Stack
Now the technology stack:
• Core engine: Java and JavaScript
• Backend server: NodeJS and other JavaScript
• Front-end server: Mobile apps based on native frameworks
• Log management: Cloud store of a small Hadoop cluster. Also,
these stored conversational logs provide data for AI/ML model
training.
• Visualization: Can be custom created using D3; if self- service is
required, then Tableau/PowerBI integration with the logs.
• Search: Elastic Search to search the conversation logs.
Maintenance
There are two critical streams in terms of maintenance scope:
 AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY
(Approved by A.I.C.T.E., New Delhi & Permanently Affiliated to J.N.T.U-GV, Vizianagaram)
NAAC “B++” Accredited Institute
Cherukupally (Village), Near Tagarapuvalasa Bridge, Vizianagaram (Dist) -531162.
www.aietta.ac.in, [email protected]

• The technology: Application uptime will be maintained with the

help of on-premise engineers and an on-call channel for all queries.
• The AI/ML brain: This will be done offsite by a team of data
scientists and updates will be pushed to the systems when ready.