lOMoARcPSD|35604100

Seminar merged

Big Data and Analytics (JECRC University)

Scan to open on Studocu

Studocu is not sponsored or endorsed by any college or university

Downloaded by Shaik Amer (shaikamer1616@gmail.com)
 lOMoARcPSD|35604100

A
Seminar Report
On
“AI Chatbot to answer FAQs”

Submitted

In partial fulfillment of the requirement for the award of degree of

Bachelor of Technology
In
Information Technology

(Session 2022-23)

Submitted to: Submitted by:

Dr. Smita Agarwal Ananya Jain
Head of Department 19EJCIT017
VII Semester (A)
Ms. Preeti Sharma
Seminar coordinator
Assistant Professor

Department of Information Technology

Jaipur Engineering College & Research Centre, Jaipur
Rajasthan Technical, University
DECEMBER-2022

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

CERTIFICATE

This is to certify that the report of the seminar submitted is the outcome of the seminar work
entitled “AI Chatbot to answer FAQs” carried out by Ananya Jain bearing Roll No.: 17 and
Enrollment No.: 19EJCIT017 carried under my guidance and supervision for the award of
Degree in Bachelor of Technology of Jaipur Engineering College & Research Centre, Jaipur
(Raj.), India during the academic year 2020-21.
To the best of the my knowledge the report

i) Embodies the work of the candidate.

ii) Has duly been completed.

iii) Fulfills the requirement of the ordinance relating to the bachelor of technology degree of the
Rajasthan technical University and

iv) Is up to the desired standard for the purpose of which is submitted.

_______________ _______________ _______________

Dr. Smita Agarwal Mr. Preeti Sharma Ms. Preeti Sharma
Head of Department Seminar Coordinator Seminar Guide
Assistant Professor Assistant Professor
Information technology Information Technology Information Technology
JECRC, Jaipur JECRC, Jaipur JECRC, Jaipur

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

DECLARATION

I hereby declare that the report entitled “AI Chatbot to answer FAQs” has been carried out and
submitted by the undersigned to the Jaipur Engineering College & Research Centre, Jaipur
(Rajasthan) in an original work, conducted under the guidance and supervision of Ms. Preeti
Sharma.

The empirical findings in this report are based on the data, which has been collected by me. I
have not reproduced from any report of the University neither of this year nor of any previous
year.

I understand that any such reproduction from an original work by another is liable to be punished
in a way the University authorities’ deed fits.

Place: Jaipur Ananya Jain

Date: 26/11/2022 19EJCIT017

ii

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

PREFACE

Bachelor of Technology in Information Technology is the Rajasthan Technical University course

(Approved by AICTE) having duration of 4 years. As a prerequisite of the syllabus every student
on this course has to make a report on seminar lab in order to complete his studies successfully.
And it is required to submit the report on the completion of it.

The main objective of this report is to create awareness regarding the application of theories in
the practical world of Information Technology and to give a practical exposure of the real world
to the student.

I, therefore, submit this seminar report on “AI Chatbot to answer FAQs”, which was undertaken
at JECRC, Jaipur. I feel great pleasure to present this seminar report.

iii

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

ACKNOWLEDGEMENT

“Any serious and lasting achievement or success, one can never achieve without the help,
guidance and co-operation of so many people involved in the work.”

It is my pleasant duty to express my profound gratitude and extreme regards and thanks to Mr.
Arpit Agarwal Director of JECRC, Dr. V.K. Chandna Principal of JECRC, Dr. Smita
Agarwal gave me an opportunity to take this seminar report.

I am indebted towards my supervisors who have allotted this seminar and his precious time and
advice during the period, which is imminent to the report.

I would like to express deep gratitude to Dr. Smita Agarwal, Head of Department (Information
Technology), Jaipur Engineering College & Research Centre, Jaipur (Rajasthan) with whose
support the seminar report has been made possible.

Last but not the least, I am heartily thankful to my friends and all those people who are involved
directly or indirectly in this seminar report for encouraging me whenever I needed their help in
spite of their busy schedule.

Ananya Jain

(19EJCIT017)

iv
Downloaded by Shaik Amer (shaikamer1616@gmail.com)
 lOMoARcPSD|35604100

Table of Content
S.no Title Page Number
College Certificate i
Candidate’s Declaration ii
Preface iii
Acknowledgment iv
Abstract vii
1 Introduction 1
1.1 Types of Chatbots 2
1.2 Applications of Chatbots 3
1.3 Architecture of Chatbot 5
2 Background Study 8
2.1 History 8
2.2 Technology Progress 10
3 Methodology 14
3.1 How do Chatbots work? 14
3.2 Building Chatbots 14
3.3 What is NLU? 20
3.4 What is NLP? 21
3.5 How to make Chatbot: Two Approches 23
4 Result 24
Discussion 30
Conclusion 32
Future Scope 33
References 35
Mapping with Program Outcomes (PO) 36

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

Figure Index

S.no Title Page Number

3.1 Component of Chatbot 15
3.2 Classification Method 17
3.3 Neural Network 19
3.4 NLP Steps 21

vi

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

ABSTRACT

Context: ChatBot can be described as software that can chat with people using artificial
intelligence. These software are used to perform tasks such as quickly responding to users,
informing them, helping to purchase products and providing better service to customers. In this
paper, we present the general working principle and the basic concepts of artificial intelligence
based chatbots and related concepts as well as their applications in various sectors such as
telecommunication, banking, health, customer call centers and e-commerce. Additionally, the
results of an example chabbot for donation service developed for telecommunication service
provider are presented using the proposed architecture.

Method: I have developed a theoretical and practical framework to assist software with a set of
Multi-Criteria Answering Chatbot in software production. This study presents a decision model
for a Chatbot answering FAQs various platform, selection problem to capture knowledge
regarding such platforms and concepts systematically.

Conclusion: The use of chatbots evolved rapidly in numerous fields in recent years, including
Marketing, Supporting Systems, Education, Health Care, Cultural Heritage, and Entertainment.
In this paper, we first present a historical overview of the evolution of the international
community’s interest in chatbots. Next, we discuss the motivations that drive the use of chatbots,
and we clarify chatbots’ usefulness in a variety of areas. Moreover, we highlight the impact of
social stereotypes on chatbots design. After clarifying necessary technological concepts, we
move on to a chatbot classification based on various criteria, such as the area of knowledge they
refer to, the need they serve and others. Furthermore, we present the general architecture of
modern chatbots while also mentioning the main platforms for their creation. Our engagement
with the subject so far, reassures us of the prospects of chatbots and encourages us to study them
in greater extent and depth.

vii

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

CHAPTER 1
INTRODUCTION

The industrial revolution 4.0 had an impact on the urgency of the education field to be able to keep
up with these developments, which later brought the term Education 4.0 . In embodying Education
4.0, one of the most required abilities from educators and educational practitioners is to be able to
integrate modern technology in their teaching. The rapid development of smartphone technology,
social media, and artificial intelligence (AI), provide challenges for educational practitioners to
utilize these technologies in developing advanced learning media. In recent decades, artificial
intelligence utilization to develop applications is massively conducted, and its products used in
almost every aspect of our life. This type of communication which occurs through digital
technology rather than in person is called computer-mediated communication (CMC). CMC forms
include instant messaging, email, chat rooms, online forums, social networks, and chatbot or
chatterbot. Chatbot is a computer program or artificial intelligence which carries out conversations
through audio or text, and interacts with users in a particular domain or topic by giving intelligent
responses in natural language. Chatbots for general purposes and for educational purposes have
been developed. However, despite chatbots’ unlimited possibility to enhance language teaching and
learning, the concept of chatbot including its advantages as a language learning medium is not yet
widely known. Therefore, the purpose of this study is to analyze the types of artificial intelligence
in the form of chatbots and the possibility of their use as a language learning medium. This study
also aims to observe a chatbot constructed as a Japanese language learning medium developed by
the author and team, and reports its’ results as an inquiry to find out further.
Chatbots are not a recent development. They are simulations that can understand human language,
process it, and interact back with humans while performing specific tasks. For example, a chatbot
can be employed as a helpdesk executive. Joseph Weizenbaum created the first chatbot in 1966,
named Eliza. It all started when Alan Turing published an article named “Computer Machinery and
Intelligence” and raised an intriguing question, “Can machines think?” ever since, we have seen
multiple chatbots surpassing their predecessors to be more naturally conversant and technologically
advanced. These advancements have led us to an era where conversations with chatbots have
become as normal and natural as with another human. Today, almost all companies have chatbots to
engage their users and serve customers by catering to their queries. We practically will have

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

chatbots everywhere, but this doesn’t necessarily mean that all will be well-functioning. The
challenge here is not to develop a chatbot but to develop a well-functioning one. Today, almost all
companies have chatbots to engage their users and serve customers by catering to their queries. We
practically will have chatbots everywhere, but this doesn’t necessarily mean that all will be well-
functioning. The challenge here is not to develop a chatbot but to develop a well-functioning one.

A chatbot is a program that communicates with you. It is a layer on top of, or a gateway to, a
service. Sometimes it is powered by machine learning (the chatbot gets smarter the more you
interact with it). Or, more commonly, it is driven using intelligent rules (i.e. if the person says this,
respond with that). The services a chatbot can deliver are diverse. Important life-saving health
messages, to check the weather forecast or to purchase a new pair of shoes, and anything else in
between. The term chatbot is synonymous with text conversation but is growing quickly through
voice communication. The chatbot can talk to you through different channels; such as Facebook
Messenger, Siri, WeChat, Telegram, SMS, Slack, Skype and many others. Consumers spend lots of
time using messaging applications (more than they spend on social media). Therefore, messaging
applications are currently the most popular way companies deliver chatbot experiences to
consumers.

Chatbots are also known as chatter bots. A chatbot is developed with the help of an (Artificial
Intelligence) application. It gives permission to humans to interact with digital devices as if they
were communicating with a human being. AI chatbots employ machine learning, an AI capability
that allows bots to become smarter over
time as they are used. AI chatbots are a natural match for customer service because of this. Chatbots
use digital instant messenger to communicate with people and can be integrated into a variety of
applications and websites.

1.1 Types of Chatbots

There are many types of chatbots available. A few of them can be majorly classified as follows:
● Text-based chatbot: In a text-based chatbot, a bot answers the user’s questions via a text
interface.
● Voice-based chatbot: In a voice or speech-based chatbot, a bot answers the user’s questions via a
2

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

human voice interface.

There are mainly two approaches used to design the chatbots, described as follows:

● In a Rule-based approach, a bot answers questions based on some rules on which it is trained on.
The rules defined can be very simple to very complex. The bots can handle simple queries but fail
to manage complex ones.

● Self-learning bots are the ones that use some Machine Learning-based approaches and are
definitely more efficient than rule-based bots. These bots can be further classified into two types:
Retrieval Based or Generative.

There are many types of chatbots available, depending on the complexity. A few of them can be:

● Traditional chatbots: They are driven by system and automation, mainly through scripts with
minimal functionality and the ability to maintain only system context.

● Current chatbot: They are driven by back-and-forth communication between the system and
humans. They have the ability to maintain both system and task contexts.

● Future chatbot: They can communicate at multiple levels with automation at the system level.
They have the ability to maintain the system, task, and people contexts. There is a possibility of
introducing master bots and eventually a bot OS.

1.2 Top Applications of Chatbots

1. Virtual reception assistant - customers to the right department automatically. It answers to the
needs of the customer 24*7, without human intervention. Using the virtual receptionist service,
you’ll be also able to track & monitor customer calls to provide a superior customer experience &
increase branVirtual reception assistant - Virtual Receptionist is a smart IVR system which helps
connect d loyalty.

2. Virtual help desk assistant - A virtual assistant is an administrative professional who works
3

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

remotely for a client or set of clients, handling appointment setting, inbound and outbound phone
calls, scheduling, travel arrangements, email, typing, light bookkeeping, and more.

3. Virtual tutor or teacher - A virtual tutor is an online program that simulates a human tutor. We're
not talking about another human at the end of a Zoom call, which can be both clunky and
expensive. Instead, a virtual tutor uses technology to identify your child's learning needs.

4. Virtual driving assistant - This tempts the driver to use their mobile phones to check incoming
messages and notifications or for communicating with someone. Hence, providing a virtual in-
vehicle assistant to drivers to interact will give them the sense of companionship during their
journey.

5. Virtual email, complaints, or content distributor - Virtual email uses your domain name in the
address (example yourname@yourdomain.com). The virtual email address is set up to forward to a
'real' email address such as your existing Golden West email account. Virtual email addresses are
often easier to remember.

6. Virtual home assistant [example: Google Home] - Virtual assistants are typically
cloud-based programs that require internet-connected devices and/or applications to work. Three
such applications are Siri on Apple devices, Cortana on Microsoft Devices and Google Assistant on
Android devices. There are also devices dedicated to providing virtual assistance.

7. Virtual operations assistant [example: Jarvis from the movie Iron Maiden] - A virtual assistant is
an independent contractor who provides administrative services to clients while operating outside of
the client's office. Virtual Operations Managers and Virtual Operations Assistants. Virtual
Operations Managers work closely with teams and business owners to ensure that company
workflows, systems, and practices are running smoothly and within budgetary.

8. Virtual entertainment assistant [example: Amazon Alexa] - A virtual assistant is a remote

employee who offers administrative support for you and your business, usually part-time. They can
do tasks that an executive assistant would typically handle, such as scheduling appointments,
making phone calls, arranging travel, or organizing emails.
4

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

9. Virtual phone assistant [example: Apple Siri] - A virtual assistant is a remote employee who
offers administrative support for you and your business, usually part-time. They can do tasks that an
executive assistant would typically handle, such as scheduling appointments, making phone calls,
arranging travel, or organizing emails.

10. Assist the visually impaired person in describing the surroundings - DO give a clear word
picture when describing things to an individual with vision loss. Include details such as color,
texture, shape and landmarks. DO touch them on the arm or use their name when addressing them.
This lets them know you are speaking to them, and not someone else in the room.

1.3 Architecture of ChatBot

Typical chatbot architecture should consist of the following:

 Chat window/session/front end application interface
 The deep learning model for Natural Language Processing [NLP]
 Corpus or training data for training the NLP model
 Application Database for processing actions to be performed by the chatbot

Following are the key components of a conversational chatbot architecture:

1. Environment
2. Question and Answer System
3. Plugins/Components
4. Node Server / Traffic Server
5. Front-end Systems

1. Environment
This is where the core Natural Learning Process (NLP) engine and context interpretation happens.
NLP Engine- NLP Engine is the core component that interprets what users say at any given time
and converts the language to structured inputs that system can further process. Since the chatbot is
domain specific, it must support so many features. NLP engine contains advanced machine learning
algorithms to identify the user’s intent and further matches them to the list of available intents the
5

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

bot supports.
NLP Engine further has two components:

 Intent Classifier: Intent classifier takes user’s input identifies its meaning and relates back to one
of the intents that the chatbot supports.
 Entity Extractor: Entity extractor is what extracts key information from the user’s query.

Agent for Dialogue Management- It manages the actual context of the dialogue. For example, the
user might say “He needs to order ice cream” and the bot might take the order. Then the user might
say “Change it to coffee”, here the user refers to the order he has placed earlier, the bot must
correctly interpret this and make changes to the order he has placed earlier before confirming with
the user. Dialog management plugin enables us to do this.

Dialogue management further has following key plugins:

Feedback Mechanism: Here the agent takes the feedback from user time to time to learn if the bot is
doing fine with the conversation and the user is satisfied with the bot’s response. This reinforces the
bot to learn from mistakes and corrects itself in future conversations.
Policy Learning: Policy learning is a higher-level framework that teaches the bot to take more of
happy paths during the conversation to improve overall end-user satisfaction.
Broadly it creates the network of happy paths and routes the conversation to end-user satisfaction.
The bot then tries to learn from the interactions and follows the interaction flow about the
conversation it had with similar users in the past.

2. Question and Answer System

This is the key component in answering users’ frequently asked questions. Q & A system interprets
the question and responds with relevant answers from the knowledge base. It has the following
components:

i. Manual Training: Manual training involves the domain expert creating the list of frequently
asked users queries and map its answers. This helps the bot quickly identify the answers to the most
important questions.
ii. Automated Training: Automated training involves submitting the company’s documents like
6

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

policy documents and other Q&A type of documents to the bot and ask it to train itself. The engine
comes up with a list of question and answers from these documents. The bot then can answer with
confidence.

3. Plugins/Components
Plugins offer chatbots solution APIs and other intelligent automation components for chatbots used
for internal company use like HR management and field-worker chatbots.

4. Node Server / Traffic Server

The server that handles the traffic requests from users and routes them to appropriate components.
The traffic server also routes the response from internal components back to the front-end systems.

5. Front-End Systems
Front-end systems can be any client-facing platforms. They can be the actual chatbot interfaces that
reside in various platforms like:
 Facebook
 Slack
 Google Hangouts
 Skype for Business

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

CHAPTER 2

BACKGROUND STUDY

2.1 History

In 1950, Alan Turing wondered if a computer program could talk to a group of people without
realizing that their interlocutor was artificial. This question, named Turing test, is considered by
many to be the generative idea of chatbots (Turing, 1950). The first chatbot with ELIZA name was
constructed in 1966. ELIZA simulated a psychotherapist’s operation, returning the user’s sentences
in the interrogative form Weizenbaum (1966). Its ability to communicate was limited, but it was a
source of inspiration for the subsequent development of other chatbots (Klopfenstein et al., 2017).
ELIZA uses pattern matching and a response selection scheme based on templates (Brandtzaeg &
Følstad, 2017). A drawback of ELIZA is that its knowledge is limited, and therefore, it can discuss
only in a particular domain of topics. Also, it cannot keep long conversations and cannot learn or
discover context from the discussion.

In 1972, PARRY appeared; It acted as a patient with schizophrenia (Colby et al., 1971). PARRY is
considered more advanced than ELIZA is as it is supposed to have a “personality” and a better
controlling structure. It defines his responses based on a system of assumptions and “emotional
responses” activated by the change of weights in the user’s utterances (Colby et al., 1972). PARRY
was used in an experiment in 1979 when five psychiatrist judges interviewed by teletype a patient to
decide whether he was a computer program or a real schizophrenic patient. Therefore, psychiatrists
gave ten diagnoses. The first psychiatrist gave two correct diagnoses; another gave two incorrect
ones. The third considered that both subjects were real patients, and the other two diagnosed that
both subjects were chatbots (Heiser et al., 1979). However, the sample of five psychiatrists is small,
and the meaning of the findings is not clear as people with schizophrenia have a degree of
incoherence in their speech. In general, PARRY is considered a chatbot with low capabilities
concerning language understanding and the ability to express emotions. It also has a low speed of
responding, and it cannot learn from the conversation.

Artificial Intelligence is firstly used in the domain of the chatbots with the construction of
Jabberwacky in 1988 (Jabberwacky, 2019). Jabberwacky was written in CleverScript, a language
8

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

based on spreadsheets that facilitated the development of chatbots, and it used contextual pattern
matching to respond based on previous discussions. Still, Jabberwacky cannot reply to high speed
and work with a massive number of users (Jwala, 2019).

The term Chatterbot was first mentioned in 1991. It was a TINYMUD (multiplayer real-time virtual
world) artificial player, whose primary function was to chat. Many real human players seemed to
prefer talking to Chatterbot than a real player. The Chatterbot succeeded because, in the TINYMUD
world, players assumed that everybody was a human and might cause doubts only if it made a
significant mistake (Mauldin, 1994). Dr. Sbaitso (Sound Blaster Artificial Intelligent Text to Speech
Operator) (Dr. Sbaitso, 2019), a chatbot created in 1992, was designed to display the digitized
voices the sound cards were able to produce. It played the role of a psychologist without any sort of
complicated interaction (Zemčík, 2019).

Another step forward in the history of chatbots was the creation, in 1995, of ALICE (Artificial
Linguistic Internet Computer Entity), the first online chatbot inspired by ELIZA (Wallace, 2009).
ALICE was based on pattern-matching, without any actual perception of the whole conversation
(Marietto et al., 2013) but with a discussion ability on the web that allowed longitude and included
any topic. However, a few years had to pass before it was improved to win the title of the Loebner
Prize of the best human-like computer program (Bradeško & Mladenić, 2012). ALICE was
developed with a new language created for this purpose, Artificial Intelligence Markup Language
(AIML), which is the most critical difference between ALICE and ELIZA. ALICE’s Knowledge
Base consisted of about 41,000 templates and related patterns, a vast number comparing to ELIZA
that had only 200 keywords and rules (Heller et al., 2005). However, ALICE did not have
intelligent features and could not generate human-like answers expressing emotions or attitudes.

In 2001, there was a real evolution in chatbot technology with the development of SmarterChild
(Molnár & Zoltán, 2018), which was available on Messengers like America Online (AOL) and
Microsoft (MSN). It was the first time that a chatbot could help people with practical daily tasks as
it could retrieve information from databases about movie times, sports scores, stock prices, news,
and weather. This ability marked a significant development in both the machine intelligence and
human–computer interaction trajectories as information systems could be accessed through
discussion with a chatbot.
9

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

The development of Artificial Intelligence chatbots went one step further with the creation of smart
personal voice assistants, built into smartphones or dedicated home speakers, who understood voice
commands, talked by digital voices, and handled tasks like monitoring home automated devices,
calendars, email and other. Apple Siri (Siri), IBM Watson (Watson Assistant |IBM Cloud, 2020),
Google Assistant (Google Assistant, your own personal Google, 2019), Microsoft Cortana (Personal
Digital Assistant—Cortana Home Assistant—Microsoft, 2019), and Amazon Alexa (What exactly
is Alexa? Where does she come from? And how does she work?, 2019) are the most popular voice
assistants. There are also many other less famous voice assistants owing unique characteristics, but
the same core functions. They connect to the Internet and, in contrast to their predecessors, they
create quickly meaningful responses (Hoy, 2018).

2.2 Technology Progress

Siri (Siri), developed by Apple in 2010, pioneered the way for personal assistants. Users make
inquiries and conversations with it through Messengers using voice commands, and it includes
integration with audio, video, and image files. Siri makes recommendations and responds to user
requests using various internet services, while it adapts, with constant use, to the users’ language
usages, searches, and desires (Siri, 2020). Although Siri is sophisticated, it is not without
weaknesses. It requires an internet connection. It is multilingual, but there are many languages it
does not support, while navigation instructions are supported only in English. It also has difficulties
hearing the interlocutor, who has a heavy accent or in the presence of noise (Soffar, 2019).

In 2011, a chatbot called Watson (Watson Assistant |IBM Cloud, 2020) was created by IBM.
Watson could understand the natural human language well enough to win two previous champions
on the quiz competition “Jeopardy”, in which participants received some information in the form of
answers and should guess the corresponding questions. Years later, Watson enabled businesses to
create better virtual assistants. Moreover, Watson Health was designed to help doctors in healthcare
diagnose diseases. However, a drawback of Watson is that it supports only English.

Google Now (Google now, 2020), developed in 2012, was initially used to give information to the
user taking into account the time of day, location, and preferences. Google Assistant (Google
Assistant, your own personal Google, 2019), which was developed in 2016, constitutes the next

10

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

generation of Google Now. It has a more in-depth artificial intelligence with a friendlier, more
conversational interface and delivers information to users predicting their requirements. However, it
has no personality and its questions may violate the user’s privacy as it is linked directly to their
Google Account.

Microsoft designed a personal assistant Cortana developed in 2014 (Personal Digital Assistant—
Cortana Home Assistant—Microsoft, 2019). It recognizes voice commands and performs tasks such
as identification of time and position, support people-based reminders, send emails and texts, create
and manage lists, chitchat, play games, and find information the user requests. The major drawback
of Cortana that has been reported is that it can run a program that will install malware (Cortana
security flaw means your PC may be compromised, 2018).

The same year, Amazon (What exactly is Alexa? Where does she come from? And how does she
work?, 2019) introduced Alexa, which is built into devices for home automation and entertainment
and making in this way the Internet of Things (IoT) more accessible to humans. An innovation is
that developers can use Alexa Skills Kit (ASK) to create and publish free or paid Alexa skills. As
we report in Section 7, Alexa introduces security issues.

Although personal voice assistants enable voice communication with their users, misunderstandings
often occur, as they cannot understand the particular language people use in oral speech or fail to
understand the whole context of the conversation.

Early in 2016, an evolution of Artificial Intelligence Technology occurred that changed

dramatically the way people communicate with manufacturers. Social media platforms allowed
developers to create chatbots for their brand or service to enable customers to perform specific daily
actions within their messaging applications. At the end of 2016, 34.000 chatbots covered a wide
range of uses (Wizu, 2018) in fields like Marketing, Supporting Systems, Health Care,
Entertainment, Education, and Cultural Heritage. Thousands of text-based chatbots with specific
features were developed for popular messaging platforms, industrial solutions, and research (Dale,
2016). Moreover, the Internet of Things (IoT) introduced a new era of connected smart objects
where the use of chatbots improved communication between them (Kar & Haldar, 2016).

11

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

Worth mentioning is also the Microsoft XiaoIce, which is an AI chatbot that satisfies the human
need for sociability. Apart from its personality, its contribution to the development of the chatbots is
that it has intelligent and emotional quotient (IQ–EQ). It establishes long emotional relationships
with its users, taking into account the cultural peculiarities and ethical issues (Zhou et al., 2019).

The way chatbots nowadays engage in a discussion is entirely different from their predecessor
Eliza. They can share personal thoughts and family drama events, be relevant but also confusing,
and deceive just as humans do (Shah et al., 2016).

A growing increase in the use of chatbots was observed, especially after 2016 (Fig. 1). According to
Fig. 2, the country that has shown more research interest in chatbots is the USA, while the United
Kingdom and Japan follow with less than one-third of the number of papers published in the USA.
Chatbots that offer services like booking services in restaurants, airlines, or searches in FAQ
without being a friendly companion, belong to Interpersonal chatbots. Intrapersonal chatbots are
close companions that live in the user’s domain and understand his needs. They are usually
connected to messenger applications like Slack and WhatsApp. Finally, Inter-agent chatbots
provide communication with other chatbots. Alexa and Cortana are two chatbots that were
integrated to communicate with each other (Nimavat & Champaneria, 2017).

The primary Goal a chatbot aims to achieve classifies them in Informative, Chat-
based/Conversational, and Task-based chatbots. When users communicate with a chatbot to get
specific information stored in a fixed source, Informative chatbots like Guardian, Facebook M, or
FAQ chatbots are used. Chat-based/Conversational chatbots hold a natural conversation with the
user like a real person would do. Finally, Task-based chatbots handle different functions, such as
room booking, and are excellent at requesting information and responding to the user appropriately
(Kucherbaev et al., 2018)(Nimavat & Champaneria, 2017).

The Response Generation method separates chatbots into Rule-based, Retrieval based, and
Generative based chatbots, which we analyze in the next section (Hien et al., 2018).

12

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

In some cases where more flexibility is needed, the operation of a chatbot can be combined with
human intervention. A Human-mediated chatbot utilizes human computation in at least one part of
it. Fully autonomous chatbots may have weaknesses that can be overcome by staff working to
integrate their intelligence into them. However, human computation lacks speed in information
processing, and it is inevitable to cope with a vast amount of user requests (Kucherbaev et al.,
2018).

Depending on the Permissions provided by the development platforms, chatbots can be divided into
Open-source or Commercial, which we further discuss in Section 9.

Finally, another classification depends on the Communication channel that chatbots utilize, which
can be text, voice, image, or all of them.

13

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

CHAPTER 3
METHODOLOGY

3.1 How do Chatbots Work?

Bots are made for a specific reason. A store would most likely want chatbot services that assists you
in placing an order, while a telecom company will want to create a bot that can address customer
service questions.

There are two categories of chatbots: one that works by following a series of rules, and another that
uses artificial intelligence.

1. Rule-based chatbots
A rule-based bot can only comprehend a limited range of choices that it has been programmed with.
Predefined rules define the course of the bot’s conversation. Rule-based chatbots are easier to build
as they use a simple true-false algorithm to understand user queries and provide relevant answers.

2. AI-based chatbots
This bot is equipped with an artificial brain, also known as artificial intelligence. It is trained using
machine-learning algorithms and can understand open-ended queries. Not only does it comprehend
orders, but it also understands the language. As the bot learns from the interactions it has with users,
it continues to improve. The AI chatbot identifies the language, context, and intent, which then
reacts accordingly.

3.2 Building Chatbot

Chatbot architecture is the spine of the chatbot. The type of architecture for your chatbot depends
on various factors like use-case, domain, chatbot type, etc. Feedback Mechanism: Here the agent
takes the feedback from user time to time to learn if the bot is doing fine with the conversation and
the user is satisfied with the bot’s response.
A chatbot is software that simulates human conversations. It enables the communication between a
human and a machine, which can take the form of messages or voice commands. A chatbot is
14

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

designed to work without the assistance of a human operator.

However, the basic conversation flow remains the same. Let us learn more about the critical
components of chatbot architecture:

Fig 3.1 Component of a Chatbot

1. Question and Answer System

As the name suggests, the Q&A system is responsible for answering customers’ frequently asked
questions. The question is interpreted by the Q&A system, which then replies with appropriate
responses from the knowledge base. It consists of the following elements:

Manual Training: Manual training entails the domain specialist compiling a list of commonly asked
user questions and mapping out the answers. It enables the chatbot to identify the most relevant
questions’ answers rapidly.
Automated Training: Automated training entails sending business documents to the chatbot, such as
policy documents and other Q&A type documents, and instructing it to train itself. From these
documents, the engine generates a list of questions and responses. The chatbot would then be able
to respond with confidence.
2. Environment
The environment is mainly responsible for contextualizing users’ messages using natural language
processing (NLP).

The NLP Engine is the central component of the chatbot architecture. It interprets what users are
saying at any given time and turns it into organized inputs that the system can process. The NLP
15

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

engine uses advanced machine learning algorithms to determine the user’s intent and then match it
to the bot’s supported intents list.

NLP Engine has two components:

Intent Classifier: An intent classifier maps between what a user asks and the type of action
performed by the software.
Entity Extractor: The entity extractor is responsible for identifying keywords from the user’s query
that helps determine what the user is looking for.
An NLP engine can also be extended to include feedback mechanism and policy learning for better
overall learning of the NLP engine.

Feedback Mechanism: This includes the feedback for the chatbot provided by the users. This part of
learning can be incorporated into the chatbot itself. Here, the user rates the interaction at the end of
the conversation. It encourages the bot to learn from its mistakes and improve in future interactions.
Policy Learning: Policy learning is a broad framework wherein the bot is trained to create a network
of happy paths in the conversation flow that increase overall end-user satisfaction.
3. Front-End Systems
Front-end systems are the ones where users interact with the chatbot. These are client-facing
systems such as – Facebook Messenger, WhatsApp Business, Slack, Google Hangouts, your
website or mobile app, etc.

4. Node Server / Traffic Server

It is the server that deals with user traffic requests and routes them to the proper components. The
response from internal components is often routed via the traffic server to the front-end systems.

5. Custom Integrations
With custom integrations, your chatbot can be integrated with your existing backend systems like
CRM, database, payment apps, calendar, and many such tools, to enhance the capabilities of your
chatbot.

16

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

There are three classification models that chatbots adopt to work:

Fig 3.2 Classification methods

Pattern Matchers
Bots use pattern matching to classify the text and produce a suitable response for the customers. A
standard structure of these patterns is “Artificial Intelligence Markup Language” (AIML).
A simple pattern matching example:

The machine then gives and output:

Human: Do you know who Abraham Lincoln is?

Robot: Abraham Lincoln was the US President during the American civil war.

Chatbot knows the answer only because his or her name is in the associated pattern. Similarly,
chatbots respond to anything relating it to the associated patterns. But it cannot go beyond the
related pattern. Algorithms can help for an advanced level of working.

17

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

Algorithms
A unique pattern must be available in the database to provide a suitable response for each kind of
question. A hierarchy is created with lots of combinations of patterns. Algorithms are used to
reduce the number of classifiers and create a more manageable structure.

Computer scientists call it a “Reductionist” approach- to give a simplified solution; it reduces the
problem.
Multinational Naive Bayes is the best example of the algorithm for NLP and text classification. For
instance, let’s look at the set of sentences that belong to a particular class. With new input
sentences, each word is counted for its occurrence and is accounted for its commonality. Then, each
class is assigned a score. The highest scored class is the most likely to be associated with the input
sentence.

Example of Sample Training Set:

Class: Greetings

“How are you doing?”

“Good morning”

“Hi, there!”

Sample Input Sentence Classification:

Input: “Hello, good morning.”

Term: “Hello” (no matches)

Term: “Good” (class: Greetings)

Term: “morning” (class: Greetings)

Classification: Greetings (score=2)

With the help of an equation, word matches are found for the given sample sentences for each class.
The classification score identifies the class with the highest term matches, but it also has some
limitations. The score signifies which intent is most likely to the sentence but does not guarantee it
18

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

is the perfect match. The highest score only provides the relativity base.

Artificial Neural Networks

Neural Networks are a way of calculating the output from the input using weighted connections,
which are computed from repeated iterations while training the data. Each step through the training
data amends the weights resulting in the output with accuracy.

Fig 3.3 Neural Networks

As discussed earlier here, each sentence is broken down into individual words, and each word is
then used as input for the neural networks. The weighted connections are then calculated by
different iterations through the training data thousands of times, each time improving the weights to
make it accurate.

The trained data of a neural network is a comparable algorithm with more and less code. When
there is a comparably small sample, where the training sentences have 200 different words and 20
classes, that would be a matrix of 200×20. But this matrix size increases by n times more gradually
and can cause a massive number of errors. In this kind of scenario, processing speed should be
considerably high.

There are multiple variations in neural networks, algorithms as well as patterns matching code.
Complexity may also increase in some of the variations
19

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

3.3 What is NLU ?

NLU helps the chatbot understand the query by breaking it down. It has three specific concepts:

a) Entities: An entity represents keywords from the user’s query picked up by the chatbot to
understand what the user wants. It is a concept in your chatbot. E.g., ‘What is my outstanding bill?’
has the word ‘bill’ as an entity.

b) Intents: It helps identify the action the chatbot needs to perform on the user’s input. For instance,
the intent of “I want to order a t-shirt” and “Do you have a t-shirt? I want to order one” and “Show
me some t-shirts” is the same. All of these user’s texts trigger a single command giving users
options for t-shirts.

c) Context: It isn’t easy to gauge the context of the dialogue for an NLU algorithm because it does
not have the user conversation history. It means that it will not remember the question if it receives
the answer to a question it has just asked. For differentiating the phases during the chat
conversation, its state should be stored. It can either flag phrases like “Ordering Pizza” or
parameters like “Restaurant: ‘Dominos'”. With context, you can easily relate intents without any
need to know what was the previous question.

20

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

3.4 What is NLP ?

Natural Language Processing (NLP) chatbot takes some steps to convert the customer’s text or
speech into structured data to select the related answer. Some of the Natural Language Processing
steps are:

Fig 3.4 NLP Steps

 Sentiment Analysis: With this, the algorithm tries to interpret the sentiment of the user’s query by
reading into the entities, themes, and topics.
 Tokenization: The NLP divides a string of words into pieces or tokens. These tokens are
linguistically symbolic or are differently helpful for the application.
 Named Entity Recognition: The chatbot program model looks for categories of words, like the
name of the product, the user’s name or address, whichever data is required.
 Normalization: The chatbot program model processes the text to find common spelling mistakes or
typographical errors in the user’s intent. It gives a more human-like effect of the chatbot to the

21

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

users.
 Dependency Parsing: The chatbot looks for the objects and subjects- verbs, nouns and common
phrases in the user’s text to find dependent and related terms that users might be trying to convey.
Like most applications, the chatbot is also connected to the database. The knowledge base or the
database of information is used to feed the chatbot with the information required to give a suitable
response to the user.

The information about whether or not your chatbot could match the users’ questions is captured in
the data store. NLP helps translate human language into a combination of patterns and text that can
be mapped in real-time to find appropriate responses.

It’s impossible to build a chatbot from scratch without knowing their main types and how they
differ. This section of the article explains what AI chatbot development is and how it can benefit
your business.

RULE-BASED CHATBOTS
Rule-based chatbots are the most basic solutions used for answering simple questions. Users
interact with such bots by clicking on predefined questions that lead to the desired answer. The
chatbot provides answers from the decision tree.

If you want an example, take a look at Facebook Messenger. The platform allows businesses to
perform automated customer support by providing buttons with possible inquires and automatically
providing answers.

While such chatbots are comparatively easy to build, they are prone to providing wrong answers
and are quite limited in functionality. In some cases, they can frustrate customers by providing
wrong answers.

AI-BASED CHATBOTS
AI-based chatbots are also called conversational chatbots or natural-processing chatbots. Such bots
rely on Artificial Intelligence chatbot algorithms and machine learning to process user inputs and
provide highly personalized answers relevant to the content.

Interaction with AI-based chatbots happens by writing questions in the human language. The bot is
able to identify different questions written by a human.

After a user types in a question, the bot identifies the user’s intent, asks additional questions if
needed, and provides answers using the dialogue tree to specify what problems a user has and how
they can be addressed.

22

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

3.5 How to make Chatbot: Two Approaches

Chatbot development can be approached in two different ways. Let’s explore how to make a chatbot
to meet all your business requirements.

1. Do it all by yourself
Today, you can create a chatbot free of charge if you do all the work yourself. This approach is
suitable if you have the needed technical knowledge and require the simplest solution.

There are numerous chatbot development platforms that require a different level of technical
expertise. Some are easier to use while others are more complicate although they provide a wider
range of features.

2. Hire a professional chatbot development company

If you want to know how to build a chatbot from scratch, you will need to look for a professional
company specialized in chatbot architecture design and development. When working with a
professional company, you get a number of benefits:

 You will work with a company that knows how to make an AI chatbot
 You don’t need to have deep technical knowledge
 You can create a chatbot designed for your specific business needs
 You are not limited by the limitations of chatbot development platforms

23

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

CHAPTER 4
RESULTS

4.1 Analysis of the Results

As the name implies, FAQ chatbots aim to answer users’ common and frequently asked questions.
Different businesses can integrate FAQ chatbots in various ways for both employees and customers,
for example:

For employees
FAQ chatbots can benefit employees by:

Facilitate employee self-service

Depending on the department, new employees can reach information about holidays, working
hours, weekly, monthly, and annual meeting times. Additionally, FAQ chatbots can provide
information about company software downloads and updates, as well as answer questions about
common IT issues.

Reduce the load on customer service employee

By implementing FAQ chatbots on business websites, customers can find responses to most of their
questions without needing to contact a live agent. By doing so you can augment your workforce
since they do not need to focus on time consuming tasks.

For customers
FAQ chatbots are excellent candidates for helping customers and improving their experience. Their
use cases include:

Help pages
Chatbots are online 24/7, respond immediately, and can manage multiple users at the same time.
Integrating chatbots to existing help pages on company websites facilitates:
 recognizing different forms of questions

24

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

 searching for answers when users are distressed or in a hurry

 updating the knowledgebase underlying the help page
Capturing leads
FAQ chatbots are a great candidate to understand customer needs by identifying their queries and
behavior. Throughout the conversation, the chatbot can ask users for their contact information and
redirect this information to the marketing department for more personalized contact.

Cross selling
Although the goal of FAQ chatbots is to answer user’s questions, AI-enabled FAQ chatbots can
further lead the conversation according to users’ previous inquiries, and provide them with info and
promotions about products or services.

In general, integrating FAQ chatbots can facilitate business workflows and provide insights about
users for product management, customer success and marketing departments. For example, the
increase in frequency of a question can indicate that users are having a hard time accessing a new
feature after a UX change.

1. Faster internal processes

Chatbots can be used to improve internal communication and processes within the company.
Chatbots could be used in the onboarding process, for example, where the new employee asks the
chatbot and gets an answer immediately, rather than having to contact various departments.

IBM reports that 72% of employees don’t really understand the company’s operational strategy. A
chatbot could be useful in answering employee questions about task prioritization, for instance.
Chatbots can also speed up and streamline internal communications for simple queries, such as
asking for a WiFi password, whether the company offers paid vacation or continuity protocol in
case a laptop crashes.

2. Cost savings
Companies’ need for growing the customer service department can be managed by rolling out
increasingly capable bots that can handle more and more complex queries.

25

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

The implementation of chatbots will incur a certain amount of initial investment costs. However, in
the long run, this cost can be lower when compared to a customer service’s representative salary,
training costs, and so on. .

Taken as a whole, chatbots’ cost saving potential make them an alluring addition to any enterprise.
Research has found out that the cost savings from using chatbots in the banking industry was
estimated to be at $209M in 2019, and will reach $7.3B globally by 2023.

Moreover, and except for the initial implementation outlay, security maintenance, performance
updates, and bug fixes, chatbots do not usually incur anything more.

3. Increased sales
Business leaders claim that sales chatbots have increased their companies’ sales by 67%, on
average. There are a couple of reasons for that:

Personalized recommendations
One reason is that bots can be a frictionless platform for presenting users with algorithm-driven,
personalized recommendations of a company’s new products and services.

24/7 availability
Bots can also boost sales, because of their 24/7 availability and fast responses rate. Customers hate
to wait, and long “on-hold times” might cause them to lose interest in the purchase. Chatbots’
instant response time ensures that the customer is constantly engaged, and interacted with, through
their customer journey.

Increased customer interaction

Bots provide another channel to reach out to your customers. Bots can be leveraged to increase
customer engagement with timely tips and offers. Real-time customer communication
(conversational commerce) thanks to chatbots helps customers to find what they are looking for and
also evaluates different suggestions.
A case study indicates that a UK-based insurance company recorded 765 customer interactions
(which is recorded as a 20% increase) within 6 weeks, following the introduction of their chatbot.
26

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

You can read our article Conversational Commerce Platforms: Data-driven Benchmarking if you
are looking for the top conversational commerce platforms to initiate your conversational commerce
journey.

4. Reaching new customers

B2B and B2Bot platforms such as WeChat or Facebook Messenger are some of the most popular
messaging apps. Being continuously active on these platforms helps companies reach new
customers who may otherwise not want to reach out to the company with an email or call.

In a survey by Telus International, it was stated that 38 percent of millennials give feedback once a
week via social media. It was noted that the number of feedback has increased in the last 12 months.
Given that Facebook has more than 300K chatbots, chatbots seem to be a way to reach new
customers.

5. Gaining a deeper understanding of customers

Your customers could rarely get the chance to directly talk to your business. Chatbots provide your
business with detailed, actionable records of your customers’ greatest pain points, helping your
company improve its products and services. The chance of selling can be proportional to the data
provided by the consumer. And chatbots can improve the rate of customer data.

Chatbots are optimal tools for organizations to learn customer expectations. In light of the data
provided by the chatbot-customer interaction, customer-specific targets can be planned. Thanks to
chatbots, the organization can use the feedback to improve on its shortcomings.

6. 24-hour availability
According to studies, over 50% of customers expect a business to be available 24/7. Waiting for the
next available operator for minutes is not a solved problem yet, but chatbots are the closest
candidates to ending this problem. Maintaining a 24/7 response system brings continuous
communication between the seller and the customer.

Of course, this benefit is proportional to how well the bots are. Bots that are unable to serve simple
customer queries fail to add value even if they are 24/7 available. The main issue at this point is
27

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

how well the chatbots can understand and solve customer problems. Finally, highlighting 24/7
availability can create backlash when bots are down due to security issues or maintenance.

7. Scalability
An operator can concentrate on one customer at a time and answer one question. However, a
chatbot can answer thousands of questions simultaneously. Thanks to the speed of the cloud,
internet, and advanced software mechanisms, the scalability of chatbots allows them to address
numerous inquiries with minimal hassle.

8. Consistent answers
Talking to different customer service representatives of even the same entity could result in
discrepancies and inconsistencies in answers. That could have many reasons. The agent the
customer talks to might be new at their job and might not have had the best on-boarding session. Or
they could just be having a tough day at work and cannot give all their attention to the customer,
thus providing a different answer than the one the customer was expecting.

The advantage of chatbots is that they function on pre-determined frameworks and leverage their
answers from a single source of truth: the command catalog. This minimizes the possibility of
inconsistency in answers and fomentation of confusion.

9. Recorded answers
Most chatbots have the ability of recording the conversation and providing the customer with a
copy of the chat’s transcript, for further use. The chat could also get archived, and the user could be
issued a support ticket for it. So if they were eventually transferred to a live agent, through the
support ticket, the customer care representative would immediately bring up the customer’s chat
history.

10. Multilingual
One of the advantages of chatbots is that they can be programmed to carry out conversation in
multiple language. This is particularly handy for global brands, operating in different markets. The
way chatbots exhibit their multilingualism is that could either ask the user, at the beginning of the
conversation, their preferred language. Or depending on where in the world the user is visiting the
28

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

company’s website from, the chatbot would automatically switch to that region’s language.

How to implement multi language chatbots

Source: Botsify
11. Endless patience
While customer reps and customers sometimes lose their patience, bots do not. The impatience of
the representative and the consumer during a conversation is one of the human-related failures.
Chatbots can show the patience that no human can provide. At this point, a human-sourced
consumer service problem can be resolved directly.

12. Convenient for introverts

Not everyone relishes constant human interactions. Chatbots give introverted users the possibility to
have their issues addressed and their questions answered without necessarily talking with a live
agent.

13. Programmability
Since chatbots function on pre-determined codes, they can be programmed to carry out various
tasks. Chatbots can arrange meetings, provide advanced search functionality, answer specific
questions, and more. As long as their command catalog is being continuously updated by
programmers, their programmability means their multi-functionality.

14. Personalization
The conversational AI capabilities of chatbots mean they can store and leverage your interaction
history with them to provide more personalized interaction. In customer service, for instance, they
could remember the customer’s name and their ticket number. This means the chatbots will be able
to instantly draw up the background information of the user to resolve their issues quicker.

29

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

DISCUSSION

When testing the last prototype we got findings suggesting that the participants did not have a
problem with getting information from a chatbot instead of a human. The information that they got
was not seen as less trustworthy, this could be supported by the fact that the chatbot provided a
source for the information it gave. It has been interesting to investigate how the participants
interacted with the chatbot and how they reported on it afterwards.
Our findings have some indicators leading towards that a chatbot could be a good alternative for
acting as a helpful friend for freshmans at a new school. Still we have to stress the fact that the
chatbot was not very intelligent and that the evaluators had to adjust their language to match the
chatbots.

Because of the scope of the project we did not have time to conduct as much user testing and re-
design to the chatbot as we would have liked. This has an impact on the validity of our research.
Through the project we have touched on some theory when making the chatbot, but this should also
have a larger focus for higher validity. Even though the participants trusted the information given in
this project we cannot say that people trusts a chatbot as much as they trust a human being. There
are also biases in our project, one of them is that all the students that we included in the project
already knew a lot of the answer the prototype could provide. Another bias is that the information
the chatbot provides could be seen as “casual” and are not crucial and/or vital This could have had
an impact on the results regarding trustworthiness.
With that being said we also think that some of our findings could give some insights into how a
very small group of people think about using a chatbot to gain information in a school context.
Some of the characteristics of our chatbot was viewed as appropriate for the given context, like
“casualness” and links to where the information was gathered. If the IFI chatbot is to be furthered
developed, this could be something to draw upon.

5.1 Productivity is important

Productivity was the most frequently reported motivation; thus, the majority of chatbot users seek
quick and consistent feedback when searching for information or assistance.
This finding might reflect the use of chatbots in the customer service domain. This finding may also
reflect a general trend for users to gravitate toward immediate communication channels. The broad

30

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

adoption of private messaging platforms such as Facebook Messenger, WhatsApp, and Snapchat
reflects users’ interest in more instrumental or goal-directed communication with fewer
interruptions compared to regular communication on Facebook and Twitter.
Information has been recognized as an important category of gratification in previous U&G studies.
Yet, the typical chatbot user’s need for information may require more immediacy and interactivity
than the information needs associated with other media. This hypothesis is in line with recent
research identifying young social media users’ need for instant gratification. For example,
Brandtzaeg suggest that youths communicating with organizations through social media crave
immediate feedback and dialogue and action-oriented engagement in order to achieve a clear goal.
Other studies have highlighted the fact that many people, particularly those from Western cultures,
seek to spend time productively and may feel guilty when they waste time. Similarly, users in this
study often referred to the quick response and productivity of chatbots as key motivations for using
them. The need for productivity might be specific to certain cultures, and so it may be worth
investigating in this user group.
Instant need for informational feedback may also be related to the concept of usefulness. Usefulness
concerns the extent to which a service is perceived as beneficial by performing a specific task
quickly and reliably. For chatbots to be successful in the studied user group, they must help users
resolve a task or achieve a concrete goal in an effective and efficient manner; in other words, they
need to be easy, fast, and convenient. Also, they need to fulfill a valued productivity goal, such as
getting help or access to information on the fly.

5.2 Entertainment and social motivations motivate fewer people but are important to some
Entertainment and fun are important aspects of social relations between humans. Likewise,
entertainment and socialization may be seen as aspects of the relationship between humans and
chatbots. The need for entertainment and a sense of social relationship is also highlighted in U&G
and recent U&G studies on social media in particular and online media in general.
Many activities in our daily life involve socialization and entertainment. Consequently, Thackara
argues that systems should provide users with a social platform or sense of community to generate
good user experiences. Similarly, Monk suggests that interactive systems should be designed to
support enjoyable social interactions. Sensitivity to this need for entertainment and social relations
might be even more important in the context of chatbot design because chatbots are more
humanlike than other interactive systems.
31

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

CONCLUSION

Chatbots potentially represent a new paradigm in how people will interact with data and services in
the future. Currently, there is a lack of empirical investigations into why people use chatbots. This
study provides needed insight into the motivational factors related to use of conversational
interfaces. Its results can guide future research on this topic, which may provide new insights and
guide future design and development of chatbots.
Having a chatbot friend allows one to communicate with friends anytime and anywhere.
Communication with chatbots tends to be more free and open because chatbots are considered to
have empathetic and non-judgmental abilities. Besides being able to overcome loneliness, chatbots
can also indirectly improve human social skills in communicating with other people. The limitations
of chatbots include the potential for misunderstanding the user's intentions and the low ability to
distinguish between good and bad. It can be concluded that making friends with chatbots brings not
only benefits but also risks, just like making friends with humans. Chatbots can be an option to
make friends, especially when there are no other friends to communicate with.

32

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

Future Scope

1. Conversational AI
Conversational AI is already on the rise. In fact, the global market has a compounded annual growth
rate of 22% for 2020 through 2025. The CAGR for the 2030s is likely to be even higher. This
investment drives innovation and development in conversational AI, expanding its capabilities even
further with deep learning and improved NLP.
Innovation will likely make conversational AI more accessible and affordable, allowing wider
adoption. Additionally, training will become easier as usage and adoption grow. One of the main
issues with today’s AI chatbots is that they remain expensive to train. Refined processes will help
make sophisticated conversational versions more affordable.

2. New Applications
AI chatbots are most well-known for their use in customer service. Innovations and better
technology will open up many more applications over the next decade. Businesses, in particular,
will benefit from these advancements. For example, new AI chatbots with advanced translation
capabilities could help companies expand globally and improve international customer service.
Advanced translation AI has already been in development for years, so this specific innovation
could go mainstream shortly.
Similarly, AI chatbots will likely become more popular in human resources departments worldwide.
This is a straightforward shift from customer service to employee service bots. HR versions are also
likely to become popular soon. They may gain ground in employee training, IT help and
administrative assistance functions.

3. Growth Through Social Media

Social media may not be the first thing that comes to mind when discussing AI chatbots. However,
it plays an important role in increasing consumer awareness of these bots. In fact, Facebook’s
Messenger app featured 100,000 active bots in 2017. That number has undoubtedly grown since
then.
Some AI chatbots focus on customer service, but others are simply free-to-use varieties that people
talk to for fun. They may not be pushing the limits of AI technology, but they are crucial to shifting

33

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

consumer perceptions of this technology. For example, more sophisticated chatbots could be
launched on social media for therapy or educational applications.

4. Better Security
Cybersecurity has never been more important. Numerous types of cyberattacks skyrocketed in
2021, particularly ransomware and phishing. AI chatbots are not exempt from the threat of these
attacks. In fact, increasing consumer trust depends on improving their defenses.
AI chatbots often collect personal information, such as payment data in customer service functions.
Protection will be a top priority for these bots in the future. This becomes even more important with
voice chatbots, such as those found in smart speakers. Consumers need to know they can trust these
devices to not save or risk any sensitive information.

5. Greater Transparency
AI has gained a poor reputation with consumers over recent years due to the issue of data bias. For
example, Amazon made headlines when it shut down its hiring AI after discovering the model was
biased against female candidates. This is the result of AI training gone wrong and leading to
discrimination. Even worse, these biases are difficult to detect in conventional AI models until
they’ve already been in use for a while.
Consumers need to know that they can trust AI for these chatbots to be effective. Some may fear
substandard service based on certain personal characteristics. Others may worry that it may
endanger their personal information. Luckily, new advancements in AI development are resolving
the issue of transparency.

34

Downloaded by Shaik Amer (shaikamer1616@gmail.com)

 lOMoARcPSD|35604100

References

1. Ahmed, A., Ali, N., Aziz, S., Abd-alrazaq, A. A., Hassan, A., Khalifa, M., Elhusein, B., Ahmed,
M., Ahmed, M. A. S., & Househ, M. (2021). A review of mobile chatbot apps for anxiety and
depression and their selfcare features. Computer Methods and Programs in Biomedicine Update, 1,
100012. https://doi.org/10.1016/j.cmpbup.2021.100012
2. Apostolou, M., Keramari, D., Kagialis, A., & Sullman, M. (2021). Why people make friends: The
nature of friendship. Personal Relationships, 28(1), 4–18. https://doi.org/10.1111/pere.12352
3. Benokraitis, N. V. (2015). Marriage & families: Changes, choices, and constraints (8th ed.).
Pearson.
4. Bhirud, N., Tataale, S., Randive, S., & Nahar, S. (2019). A literature review on chatbots in
healthcare domain. International Journal of Scientific and Technology Research, 8(7), 225–231.
http://www.ijstr.org/finalprint/july2019/A-Literature-Review-On-Chatbots-In-Healthcare-
Domain.pdf
5. Bosch, M., Fernandez-Borsot, G., Miró I. Comas, A., & Figa Vaello, J. (2022). Evolving
friendship? Essential changes, from social networks to artificial companions. Social Network
Analysis and Mining, 12(1). https://doi.org/10.1007/S13278-022-00864-1
6. Brandtzaeg, P. B., Skjuve, M., Dysthe, K. K., & Folstad, A. (2021). When the social becomes non-
human: Young people’s perception of social support in chatbots. CHI Conference on Human
Factors in Computing Systems, 1–13. https://doi.org/10.1145/3411764.3445318
7. Chan, G. H. (2020). Intimacy, friendship, and forms of online communication among hidden youth
in Hong Kong. Computers in Human Behavior, 111, 106407.
https://doi.org/10.1016/j.chb.2020.106407

35

Downloaded by Shaik Amer (shaikamer1616@gmail.com)