DEVELOPMENT OF SINDHI CHATBOT FOR SHAH

ABDUL LATIF UNIVERSITY KHAIRPUR

MASTER OF SCIENCE (MS)

IN
COMPUTER SCIENCE

By
SHAHIDA
REG. NO. MS-COMP-2021-19

SUPERVISOR:
DR. ASADULLAH KEHAR

INSTITUTE OF COMPUTER SCIENCE

FACULTY OF PHYSICAL SCIENCES
SHAH ABDUL LATIF UNIVERSITY, KHAIRPUR
SINDH, PAKISTAN
2024
 AUTHOR’S DECLARATION

I hereby state that my MS thesis titled “Development of Sindhi Chabot for Shah
Abdul Latif University Khairpur” is my own work and has not been submitted
previously by me for taking any degree for this University (Shah Abdul Latif
University, Khairpur) or anywhere else in the country/world.

At any time if my statement is found to be incorrect even after my Graduate that

university has the right to withdraw my MS/MPhil/PhD degree.

Name of Student: SHAHIDA

Date __________________________
 PLAGIARISM UNDERTAKING

I Shahida, MS scholar solemnly declare that research work presented in the thesis
titled “Development of Sindhi Chatbot for Shah Abdul Latif University Khairpur” is
solely my research work with no significant contribution from any other person. Small
contribution/help wherever taken has been duly acknowledged and that complete
thesis has been written by me.

I understand the zero-tolerance policy of the HEC and university (Shah Abdul Latif
University, Khairpur) towards plagiarism. Therefore, I as an Author of the above
titled thesis declare that no portion of my thesis has been plagiarized and any material
used as reference is properly referred/cited.

I undertake that if I am found guilty of any formal plagiarism in the above titled thesis
even after award of MS degree, the University reserves the rights to withdraw/revoke
my degree and the HEC and the University has the right to publish my name on the
HEC/University Website on which names of students are placed who submitted
plagiarized thesis.

Student/Author Signature __________________

Name: SHAHIDA
 CERTIFICATE OF RESEARCH GUIDE(S)

This is to certify that the thesis titled Development of Sindhi Chatbot for Shah Abdul
Latif University Khairpur Things submitted to Shah Abdul Latif University, Khairpur
in partial fulfillment of requirement for the award of degree of Master of
Philosophy/Master of Science/Doctor of Philosophy in _Computer Science, is a
record of original research work done by Miss. SHAHIDA under my/our supervision
and guidance.

GUIDE/SUPERVISOR
Mr. /Ms. ____________________________
 CERTIFICATE OF APPROVAL

This is to certify that the research work presented in this thesis, titled “Development
of Sindhi Chatbot for Shah Abdul Latif University Khairpur” was conducted by
institute of Computer Science Under the supervision of Dr Assadullah Kehar.
No part of this thesis has been submitted anywhere else for any other degree. This
thesis is submitted to the institute/department of computer science, as partial
fulfillment of the requirements for the degree of Master of Philosophy/Master of
Science/Doctor of Philosophy in the field of computer Science, institute of Computer
Science Shah Abdul Latif University, Khairpur.
The thesis is evaluated by:

Examination Committee:

a) External Examiner 1: Name

Signature______________
(Designation & Office Address)

b) External Examiner 2: Name

Signature______________
(Designation & Office Address)

c) External Examiner 3: Name

Signature______________
(Designation & Office Address)

d) Internal Examiner:
Signature______________
(Designation & Office Address)

Supervisor Name: ____________________

Signature______________

Name of Director
Institute of Compute Science
Signature______________

Dean ____________________
Faculty of Physical Sciences
Signature______________
DEDICATION

To my parents
Teachers
friends
& my
family
 TABLE OF CONTENTS
TITLE PAGE NO.

AUTHOR’S DECLARATION
PLAGIARISM UNDERTAKING i
CERTIFICATE OF RESEARCH GUIDE(S) ii
CERTIFICATE OF APPROVAL iii
DEDICATION (optional) iv
LIST OF TABLES x
xi
LIST OF FIGURES
xii
LIST OF
xiii
ABBREVIATIONS
xiv
ACKNOWLEDGMENT
Abstract CHAPTER-1 1
Introduction 1
1.1 Background Overview 1
1.2 Area of Research 3
1.3 Problem Statement 3
1.4 Objectives 4
1.5 Significance of the Study 4
1.6 Scope of the Study 5
1.7 Thesis Organization and Structure 5
1.8 Summary 6
CHAPTER-2
Literature Review 7
2.1 Historical background of the Study 7
2.2 Summary 13
CHAPTER-3 14
Research Methodology 14
3.1 Research Design 15
3.2 Research Methodology 16
3.3 Structure of Chatbot Development 17
3.4 Data Source Collection 18
3.5 Query type of Study 19

3.6 Summary 20
CHAPTER-4 21
Results 21
4.1 Evaluation Metrics 21
4.2 Test Cases 21
4.3 Confusion Metrics 22
4.4 Summary 23
CHAPTER-5 25
Conclusion, Summary of Findings 25
5.1 Conclusion 25
5.2 Limitations 26
5.3 Future Work 27
5.4 Summary 29
References 30
 LIST OF TABLES

Table.3.1 Types of Question and Queries 19

 LIST OF FIGURES

Figure.1.1 Chatbot common in social networks 3

Figure.3.1 Proposed Research Design of the Study 15
Figure.3.2 Proposed Model of the Study 16
Figure.3.3 Structure of Chatbot Development 17
Figure.4.1 The Test Cases TP,FP,TN and FN Results 21
Figure.4.2 Confusion Metrics Results 22
 LIST OF ABBREVIATIONS

ML Machine Learning
UI User Interface
TP True Positive
TF True False
TN True Negative
NF Negative False
 ACKNOWLEDGEMENT
Thank you so much for Dr.Asadullah Kehar, my thesis supervisor's guidance and
support to my thesis. It is his professionalism, affirmative comments, and timeless
encouragement that underpin not only my thesis development but also my own. I am
surely grateful for this guidance and full effort they have devoted to me.

It is the love and unswerving care of my parents, Manzoor Ahmed shaikh, that spring
me into existence and have made me the person I am heartily thankful to them. They
have been through interventions, confidence analogies, or faithful beliefs, which are
the necessary conditions of mine success.

I will never be able to repay their infinite support and the most important lesson they
left me with, which is to knock down walls, both in my life and in my own
imagination. I graciously acknowledge the personal thanks for Faculty of Physical
Sciences at Shah Abdul Latif University, Khairpur for generously offering me the
fundamental facilities that have helped in carrying out this research. I couldn’t be
more thankful for the staff and faculty who went out of their way to help and worked
with me all the while I was learning. I would like to express my gratitude to Research
Fellows that not only provided an opportunity to collaborate, but also discussed
insightful things and were very critical putting suggestions which were valuable and
hence resulted in this work to be of better quality. Also want to add my thanks to my
friends and relatives who left their emotional and mental support for me throughout
the pain and pleasure of this academic procedure. Your constant words of
encouragement and unquestionable belief in my capabilities which have been
instrumental in keeping me energized when the worst seems to have happened. In the
last place, I am appreciable to all the parties both directly and tangentially who have
hired in the process of accomplishing this thesis. It means a lot to get your help and to
see your efforts in this project, so I thank you from the bottom of my heart.

(SHAHIDA)

ABSTRACT
This thesis provides the development of a Natural Language Processing (NLP)
chatbot, and it is mainly created for Sindhi language which is a well-known major
language by many people but is low in computational resources. The BOT project,
which is being carried out jointly with the SALU (Shah Abdul Latif University), aims
to develop a Sindhi chatbot that can comprehend and respond to user questions in a
natural conversational style and therefore substitute the need of a human like assistant
for SALU University community members.

Sindhi is in the shortage of NLP resources which in fact it has been the main barrier of
artificial development for chatbots. This project recognizes this gap and therefore, it
developed a relevant platform for SALU University. research will consider the
hardships that are associated with without the availability of functional Sindhi NLP,
for example, as is exemplified by lack of readily available data and pre-trained
models. Next, it will cover the chosen methodologies for these problems, and it will
include strategies on collecting data and the selection of NLP models.

The development of this chatbot will be done to concepts and ideas that are vital to
SALU University's operation. Such means that may involve the integration for
example of features like: What I have been focused on this week for my service-
learning hours has been answering student and faculty questions. Here come some
details on the services available at the university: information on programs, events and
how to use them. Helping with administrative functions or procurement of resources
needed.
Listen to the given audio and summarize or paraphrase it in your own words.

Among the minor technical details of the proposed chatbot architecture, the internal
system hierarchy will be briefly mentioned to understand the user queries in Sindhi as
the Natural Language Processing feature. Dialogue maintenance to keep conversation
cohesion and in context with the discussion. One of NLG technologies for presenting
users with stimulating and natural questions.
Assessment of the effectiveness of the chatbot will be determined by various
performance metrics including accuracy, fluency and user satisfaction. The thesis will
explore the potential impact of the chatbot on the SALU University community,
including Web page accessibility to the university information and services. We have
developed corpus and via the confusion matric, the entire system accuracy and
performance through the test cases have been tested and received 95%.65.

Transformed communication and outreach for Sindhi speaking students and staff
members. Possibility of used in college infrastructures. By participating in developing
Sindhi NLP technology, this project adds to the state of the art and precondition for its
using in the educational setting. It holds the potential to: Upend the digital divide gap
within SALU University by putting in place the necessary digital tools so that the
students are better able to cope with the changing trends of the education sector.
Implement instant messaging, push notification, make user-friendly, and have smooth
navigation.

Keywords: Chat Bot, Sindhi Questions Corpus, Confusion Matrix, Machine

Learning ML, NLP
 CHAPTER 1 INTRODUCTION

This chapter covers the overview of the Chat bot development which fulfils the need of
conversation regarding related questions of admissions, general queries, fees etc. this chapter
would be helpful to understand background of the research.

1.1 Background / Overview

A study is of messaging happening between an interlocutor and a chat bot which could even
be contacted from places as remote as ever day. Website based chatbot may be easily
translated into local version by only pressing a few buttons without remarking other
information.[2] Of most of the college and university information as well as counseling
service, bot via chat provides only a little that is necessary. [3] One of the chatbot's can be
found on our facilitated website which is open for everyone who can assess it. A stepwise
approach in which AI and ML will be used in lieu of this is anticipated. PHP – [4] Chatbots
are designed and written in the language used for development. The interface enables the user
to send questions on various topics about college. [5] Once the query is made and becomes an
input for the algorithm, the algorithm processes it to bring an actual result. The interface in
the system is like the one of a class or usual Live contextual messenger [6]. There are several
characteristics which might end up being 'developed' by the online learning. One of them is
having a moment to respond to a customer's question and inquiry being unavailable.
Fortunately, in this time of digital age - data should be accessed in a live performance -
heaven of interactivity. An interactive information is what most users are looking for, since it
has the potential of highly attracting them by providing exactly what they want.

Thus, researching Chatbot systems that are designated to be the marketing assistant of a
private institution university referred to as UNISEL called UNISEL Bot, is the core of this
research. The establishment of chatbot with the marketing department will permit the
adoption of inbound marketing strategy that will allow the marketing department to exchange
the ideas between the target group through the sending and answering of student queries
which deals with the future. Agile methodology was employed during the learning process of
the development of this Chatbot system. In terms of data gathering the qualitative research in
the form of interview method with students and University staffs will be applied. UNISEL

1
Bot System is delineated in four diagrams that show the system flow. UNISEL Bot system
was a collection of seven modules which comprised the Ask Question, Feedback,
Registration,

Event, Appointment, Survey, as well as the Map module. Other Chatbots might sometimes in
the future, and these can be used widely to the point that they replace the traditional method
of human customer service. They can also help in building analytic data that in turn can used
to capture user data and feedback.[7] The fact that the number of students enrolling in the
universities has increased annually and the issue of lack of counselors to meet the counseling
requirement is evident. Subsequently, “Building AI
Chatbot to support admissions and career guidance for universities” has been the researched,
analyzed and understood the problematic and challenges in AI, career counseling and
enrollment support. Methods/Statistical analysis: Creation of an online chatbot suits our
needs just perfectly.

A chatbot can become available, for the whole 365 days around the clock. In addition, chatbot
deals with requests of students and parents automatically without individualization and also
quickly and uniformly sends already-composed information regardless of a scenario.
Findings: With this in mind, the organization chooses to conduct a research study on this
issue. Then curriculum design occurrence supports functioning of vocational guidance and
advising education enrollment activities with at the same time the integration of the chatbot
into the overall school system which will be for the admissions counseling process.
Improvements/Applications: Meanwhile, the study has indeed mapped out the data
architecture about freshman orientation activity as well as transfer students and used natural
language processing along with machine learning operations to make out such gleanings.

The research will be able to impart the knowledge to work on the overcoming of the
problems and performance in university admissions consultancy [8]

2
 This Photo by Unknown Author is licensed under CC BY

Figure.1.1 Chat Bot Common in Social Networks [8]

1.2 Area of Research

The area of the research is pertaining to the natural processing language in which the chat box
of Sindhi language can understand the questions and students’ queries based on admissions
and fee etc.

1.3 Problem Statement

A chatbot means a students as well as the faculty with a conversational interface that assists
the students to fulfill many tasks related to the university such as In that case, the chatbot
should be able to solve queries, offer info on the courses, happenings, and most importantly,
keep in mind the administrative tasks such as the registration and the payment of the fees
Presently, the process of admission is a manual one and does not comprise the online virtual
assistant responsive platform for the each department of SALU Khairpur. Moreover, the
students cadre are not so helpful and they do not facilitate the student in any way while
applying in SALU because they are outside the city or newly admission candidates.

Additionally, the similar provisions would be included so that the program can speak in the
local language Sindhi. Inclusiveness within the language will enable the robot to a larger
3
extent for a wider region in the area. The main goal of the research is, of course, to be
delivered to the road by implementing AI driven automated and real-time support that helps
to cut the time on mundane tasks and improves effectiveness. The research intends to
implement a chatbot based on current language processing as well as students’ queries and
concerns of the SALU Khairpur regarding their university documentation, educational
system, and faculty.

1.4 Objectives

The main objectives of the study are:

1. To collect digital texts of Sindhi language

2. To design Sindhi NLP rules for communication between user input and Sindhi Language
Corpus.
3. To create a Chatbot panel for queries responsive.

1.5 Significance of the Study

Developing a Sindhi chatbot for Shah Abdul Latif University Khairpur can be significant for
improve accessibility by provide quick and easy access to information in Sindhi language,
making it easier for students and other users who prefer to communicate in Sindhi.
Availability of 24/7 hours, the system can be available 24/7 hours, providing users with
round-the-clock support and access to information and enhanced user experience; a well-
designed chatbot can provide an interactive and engaging experience for users, making it
easier to find information and complete tasks. By automating routine tasks, a chatbot can help
save time and reduce costs for the university.

The chatbots surely can dig up large quantities of data and give back answers fast, so the
responses are more prompt and efficient. An intelligent Sindhi chatbot incorporated with the
Shah Abdul Latif University, Khairpur can be an effective instrument in the education of the
people by making the flow of information vis-à-vis user experience and ultimately improving
the efficiency.

4
1.6 Scope of the Study

A Sindhi language Chatbot can indeed be a success for the delivery of student-oriented
services by assisting students in receiving details about available programs, entrance
formalities and other essential information. It provides faculty members with the avenue to
get acquaint with the school's policies, up-to-date research opportunities, and other necessary
details. Furthermore, it can offer student assistance and support by either answering their
questions, mentoring them and connecting to relevant sources. A chatbot serve as a resource
center for students and faculty if one asks for Library resources, procedures, and so on.
Moreover, it can function as a medium of distribution for sharing information about the
university, its courses, events which take place are other relevant details to the public.

1.7 Thesis Organization & Structure

The whole work is divided into five parts. The chapters have text, images, and a table that
represents as the documentations of the research further discussed as: The chapters have text,
images, and a table that represents as the documentations of the research further discussed as:
chapter one, which is based on introduction, problem statement, research objectives, research
scope and finally the importance of the study is the first chapter in this research article In
overview of Chapter 2, are disserted the more relevant papers to the study. Through the
research phase, chapter 3, expresses the methods and procedures used for the study. The study
in chapter 4 showed the concluding period of the research.

Chapter 5 outlines the summary of the research and suggest future, expansion and
development of the research agenda.

5
1.8 Summary

In this chapter, we have discussed various points including, background, relation between
CNN and Quantum ML, motivation, research questions, objectives, hypothesis statement,
methodology and thesis compilation part. It is a useful chapter before moving towards
understating of entire research dissertation. Subsequently, the fundamental part of this
research is model which follows the cardiovascular disease. The briefly declaration of this
disease and its ratio is given in the Figure.1.1 taken from the world health organization and
other resources as well.

6
 CHAPTER 2 RELATED

WORK

The literature review is the chapter in which we have covered all the possible perspective of
the study related to our research area.

2.1 Historical Background of the Study

This study aims to give description to the components to be added for self-driven Smart
Chabot for a website of a university. The paper is confined only to the abstract category and
takes the year of 2017 to 2020 as the according period to be search for the existent academic
modules. The research will be assigned to explore how different universities could create
their own Conversational Sector. The study reveals a list of core constituent parts that
constitute a Smart Chabot structure. Smart chatbot structure that enable quick responses,
natural language communication with the integral interface to general purpose human-
oriented systems. Nevertheless, the Sturdy does not show the components arrangements that
are one of the good designer’s mechanisms during product development and expansion. To
make the Chabot capable of storing data warehouses, a Cloud service which in this case has a
role to play is necessary [8].

The Un-Bolt alias University Chabot, it gets information of universities to its users. For a
student, applying a VR technology to a university or college website as an accessible
resource from anywhere has become very convenient. The project is based on the research of
the AI and ML definitions that are coded in the PHP language in order to create an
application. Users can inquire about university related issues. These requests are then used as
an input of the algorithm which then displays the consistent response. The design of the
project screen imitates a usual messaging application. The application purposes include just
designing the store as the data is received but in specific and necessary manner. A user's
query is pre-processed and the Chabot generates recommendations based on confined options
where the user has to make a choice. Truth be told, in keeping with such practice Unibot will
store all pieces of user’s information locally using Query as well. An algorithm that wins the
face off is chosen and it works with memory sensitive Unibot, tapping little into a database

7
with minimum hits. bot process involves the cases of all the words and their spellings. Then it
converts the texts into words. After that one an SQL query is utilized and then regular
expressions are used to load the words ... into the table that is specifically used for the words
that are highly used. If no result is formed, the Chabot the sizes an apology towards the user
at that point. The Unibot is a rapid response interactive stress reliever, which is able to reflect
the humanized comments, and recognize most of the question patterns. As such, the
plagiarism detector of Unibot provides students with the ability to do their assignments user-
friendly enough and recycle the algorithm and the design required folding the information
from any university.

Industry 4.0 plus got infected with Radiance pandemic hence the emergence of chat robots to
this end by trying to achieve two main purposes namely elimination of the vacant staff and
offering a 24/7 online support. The Marana University library in Indonesia is focused on the
idea of prototyping by automatic chatbot that makes use Dialog flow, an Artificial
Intelligence technology stage, to come up with a Chabot for Frequently Asked Questions
(FAQ). The model comprises seven stages: transcribing user requirements, system/subsystem
prototyping, prototypes assessments, prototypes improvement, system testing, system
implementation, and system maintenance. The Chabot has been successfully established
using Dialog flow, with the adding of phrases to quarter various user variations in the Intents
feature increasing its accuracy. However, the Chabot faces boundaries such as incorrect
replies when using diverse languages or by means of many abbreviations. Future
developments in AI and high totaling could make the Chabot more consistent and reduce
librarian workload. The Chabot will continue to be advanced and kept, providing book
reports, information on book copying status, new member registration, and user satisfaction
[10].

The research explores the implementation and impact of the Dinus Intelligent Assistance
(DINA) Chatbot in the context of university admission services.

Taking a look at the eroded role of technology in education, the paper defines the challenges
that traditional manner of admission used to have and offers DINA as a solution for the
smoothness and betterment of user experience. The review of DINA is discussed as the
system's core function is main subject, modifications are made, and DINA's integration with
admission systems. It cites the benefits of improved availibility, quick and easy information

8
and security issue considerations. The study browses thru the possible hindrances, solutions
to user adoption, technical limitations and the scenario of the future integrating these modules
with virtual assistants. The article is concluded by giving examples of successful adopters and
a challenge for all sectors to go for wider adoption and collaboration. However, the role of
DINAD in university admission service transformation is spelled out in its entirety, which
completes the article.

In this study, the practical application of UNIBOT, a web-based AI-enabled human-like

chatbot that interacts in a chat mode, and how it impacts the university experience will be
diversely explored. The present research delineates AI application which is centered on
designing a self-sufficient and user-friendly chatbot to interact with clients in a very casual
way. In the overview section UNIBOTs development process, significant features, and their
compatibility with existing web platforms are elaborated on. What you should outline is the
advantages of using UNIBOT. The more important ones are those like using UNIBOT to
enrich the user experience, to empower them with personalized assistance and increase the
overall accessibility. The paper deals with prospective problems and takes into consideration
the likely future direction for UNIBOT as an important mechanism enhancing the functional
features of web-based interactivity platforms. In general critical review covers the change of
AI chatbots influence on the university atmosphere, it is implied close by UNIBOT too.

The aim of the research is to execute as well as analyzes Dialogflow chatbot that will provide
services in Matana University Library. This research examines the deployment of chatbot
technology to enhance the service quality for library users and to offer an effective and
efficient way of acquiring needed information. Chatbot with Dialogflow technology enables
the respond quickly to the popular queries with the best interface for the library users. The
overview shows the main features of the chatbot, its role while providing the user with instant
answers to the frequently asked questions (FAQs), and integration with Dialogflow for the
purpose of natural language processing uses. In this the university library-Matana strives to
improve on accessibility of information, users support and the whole library utilizes will be
efficient and full of interactivity as well [13].

The study undertakes a work on constructive and operative smart chatbot educational model
which is designed for a university website. The research aim is to express and examine the
model of the University of the Third Millennium, highlighting its applicability in the

9
academic setting. Rounding off the passage underscores the convergence of more
sophisticated technological methods such as natural language processing and machine
language patterns to heighten the chatbot’s intelligence and performance. The paper intends
to create an immersive and antenatal online communication channel on the university website
by building a dynamic and responsive conversational medium. Via the different parts of the
Smart Chatbot Model for Academics, which is the main innovative solution to deal with the
multiple student academic problems and ask for academic support within the college
community [14].

The research material basically deconstructs a thesis that involves a bilingual chat bot
describing a university admission process. The functional study explains about high-level
language handling feature that helps in improving communication between the bot and the
users in both the languages. Tocquefieldnicating by linguistic abilities in the admission
context, the research tries to simplify communication and interaction between prospective
students and the university. Overall, the summary points at the practical signs of a bilingual
chatbot that would have helped address the applicants' different language capabilities, while
at the same time it was making the process of applying easier and more user-friendly.
Generally, the research leads to the development of the language assistant technologies,which
makes the admission process more effective and submodulus through application of
innovative technologies.

The study is in line with designing and implementing a conversational chatbot for the College
events and the university ones alike. In this study, we elucidate the deployment of chatbots
situated to improve the visitant experience when it comes to a university's open house event.
The avatar focuses on the interactive feature which is serves as a virtual guide furnishing all
vital campus information including the facility, academic programs, and schedule of events.
This overview highlights that interactive chatbot becomes a critical push for prospective
students and visitors for attending university open days, getting real-time guidance, and better
understanding of campus life at university open days. Thus, the research is the enhancement
of the chatbot usage that is focused on the improvement of the outreach and engagement
strategy of universities during key events.

The work deals with the creation and deployment of a spoken-language university
information chatbot interface. This study is passionately about the discussion of voice

10
recognition technology integration within the university information system to improve the
user experience and accessibility. The designed chatbot, which is voice command and query-
based, will ensure a fortuitous and problem-solving access to students for information about
university services, resources, and other relevant facts. The latter touches upon the possible
benefits of networked speech interface including the ones used for specific reasons or general
purpose. In sum, the study highlights the progression of chatbots, by adopting voice-enabled
features for more humanized and convenient interaction with students.

Developing a Facebook chatbot for university inquiries using deep learning approaches in the
detail of macro seismic, especially for a RASA framework, is the main issue that the research
focuses on. Besides, this study focuses on the incorporation of deep learning algorithms
which in turn contribute to the better text understanding that is carried out by the chatbot
resulting in precise and well-informed responses. The study gives the RASA structure the
importance it deserves, hence aims at improving the process and experience of a dialog
between the potential students and the university by means of Facebook messenger. The
summary underscores the potential benefits of employing deep learning in chatbot
development for improved accuracy and efficiency in addressing universityrelated queries.
Overall, the study contributes to the advancement of intelligent chatbot systems for university
communication and information dissemination [18].

2.2 Summary

In this chapter, a review of the literature has discussed. The most related articles were cited. A
research gap was found based on the literature. There was limited literature exists that
represents Sindhi chatbot development.

11
12
 CHAPTER 3

Research Methodology

This chapter represents the research design and methodology of the study which provides the
overall process of research conducting including research design, research methodology,
which comprises the research framework or model.

3.1 Research Design

Research design is a general outline or an overall blueprint of the study processes such as
data collection, analysis, and interpretation of the empirical findings of a given study. It is an
important part of the inquiry process because it allows one to make sure the research is
professionally done, strictly adheres to the standards of rigour, and efficiently resolves the
research questions. The investigation plan does the structuring work for researchers as a
frame of reference for them to make sound decisions using the information at hand.

13
 Figure.3.1 Proposed Research Design of the Study

Figure 3.1 illustrates the research design of the study. At the beginning of the research was to
review of the literature, during the literature a significant problem statement was found. The
research objectives were declared based on statement of the problem. Besides, research
questions were developed for data collection process, after collection of the data, Chatbot was
developed. Though, for the system evaluation process was accomplished also. After
evaluation of the system, seminar was delivered and thesis write-up done also. This is the
described view and its details have been given in the methodology section.

14
3.2 Research Methodology

Research methodology is a system of procedures that researchers work through when making
their plans and executing their studies. Whose main part is formulation an overall strategy
and plan to answer to research questions. Sample sentence: One of the most undeniably
transformative effects of AI on healthcare is its ability to improve patient outcomes. A
detailed methodology contributes to the research being exposited in a methodical, exact, and
moral way.

Figure.3.2 Proposed Model of the Study

As Figure 3.2 represents ,the Model of Sindhi Chatbot for Shah Abdul Latif
University Khairpur (SALU) which would be acting as an intelligent messenger of this

15
university at first would be allowing user to insert those quries of teachers or students of this
university and then feed it to the Sindhi NLP Model within which the rules will be made in
order to communicate with the sindhi language. The Figure.3.2 follows the structure of ML
and can be seen in the next section.

3.3 Structure of Chatbot Development

This is the structure of the chatbot development for the Shah Abdul Latif University and
through this execution and structure, entire the process have been covered.

Figure.3.3 Structure of Chatbot Development

16
Figure 3-3 demonstrates the structure of chatbot development; initially the target was to
compile the question answers of student’s common question and publicly available university
information as data collection. Rasa NLP preprocessing which consists significant part of
Rasa chatbot framework software development. Rasa is an open-source AI-powered
framework that is underpinned by medfly and so make bots for chatty. This architecture is
intended to assist developers in the creation of models for natural language understanding
(NLU) as well as dialogue management systems, and to enable them to build chatbots
intelligently and within the context. [Xin, Z., & Yi, Y. (2021)] Analytical and rasa assessment
study of open source chatbot framework.Int. Talbot & J. Eng. Res, 9(06), 1011-1014.

After that, trained the model based on the Sindhi data then created logically rules onAlthough
the current drawbacks and intricacies of quantum computing limit its widespread application
outside of academia and research, these challenges are not insurmountable. Once overcome,
quantum computing could address problems that exceed the capabilities of even the most
powerful supercomputers. These issues often involve numerous variables, permutations, and
uncertainties, and solving them could lead to significant advancements in modern supply
chains, the development of new biological treatments for human diseases, and advancements
in artificial intelligence. The above Figure is the model representation of entire architecture
based on the same as CNN but follows the images based on quantum, similarly, the text to be
vectorized has follows the pattern of operating system environment.

3.4 Data Source Collection

The data collection is the long and complex procedure however we have collected as much as
we can and the resources are given below;

• Understand the University and Its Services

• Identify Target Audience and User Needs
• Collect University-Specific Information
• Develop FAQs and Common User Queries
• Collect Local and Cultural Information
• Language and Linguistic Data
• University Policies and Procedures
• Campus Maps and Facilities Information

17
 • Events and Updates
• Contact Information
• Local Events and Announcements
• Legal and Privacy Considerations
• Gather Data in Sindhi Language

3.5 Query type of the Study

Query types of the research cover the category of the questions which stored in the Chatbot
for creating the responsive according to students need.

Table.1.1 Query Types

S.NO Query-Type Queries/Answers

1 General Queries 40

2 Admissions Queries 25

3 Fees Queries 17

University Timing i.e;

4 Class timing, semester duration, pick & drop 43
queries

Table 3.1 describes the query types of study, there are four types including general queries
which comprises 40 queries with answers, admission queries cover the 25 queries, fees

3.6 Summary

18
This chapter 3 covered the methodology of the entire research. In this chapter, the research
design, research structure, proposed model of the research, question and their types, and data
collection sections have been discussed and it would be highly recommended for those who
wants to go with this type of research.

CHAPTER 4

Result and Discussion

19
This chapter represents the results and discussion of the study. The percentage of all items
scale frequencies has been calculated. The descriptive statistics parameters including mean,
median, and standard deviations have been generated.

4.1 Evaluation Metrics

The metrics are the core way of evaluation performance via the accuracy, precision, recall, f1
score confusion matric and these are the main core parts of the entire research which we used
for calculating results [36],[37],[38],[39].

4.2 Test Cases

Since long time the developers, researchers and many other experts are using the test case
approach in which we validate each action that we actually required so we have created testes
case,

Figure.4.1 The Test Cases with TP,FP,TN, and FN

4.3 Confusion Metrics

20
We have calculated each perspective of each question including admissions, general queries,
fee related queries and calculate the true positive (TP), false positive (FP), true Negative (TN)
and false negative (FN). The complete result of each one is given in the Figure.4.2.

Figure.4.2 Confusion Metrics Results

4.4 Summary

In an era where digital communication prevails, Natural Language Processing (NLP)

technologies serve as indispensable tools for facilitating seamless interactions between
humans and machines. The Sindhi language, with its rich cultural heritage and linguistic
diversity, presents a unique challenge and opportunity for the development of NLP solutions
tailored to its nuances. This thesis focuses on the conceptualization, design, implementation,
and evaluation of a Sindhi NLP chatbot customized for the SALU (Shah Abdul Latif

21
University) community, aimed at fostering effective communication and information
dissemination in Sindhi.

The Sindhi NLP chatbot is engineered on a robust architecture comprising sophisticated NLP
components, including tokenization, part-of-speech tagging, named entity recognition, and
sentiment analysis, and meticulously fine-tuned to accommodate the intricacies of the Sindhi
language.

Utilizing the latest machine learning technologies along with deep learning models, the
chatbot has the ability to understand the user queries, produce the desired responses keeping
the context in mind and by maintaining the flow in order to have an effortless conversation in
Sindhi.

The creation process requires multifaceted data collection, preprocessing, and model training
with a wide range of text data in Talu language that has been acquired solely from SALU
library resources and scientific papers. By means of continuous refinement and experiments,
the chatbot's operations are tuned to produce the maximum precision and smoothness across
the whole-span of the SALU topics, for instance, a wide range of agricultural, veterinary, and
environmental technologies.

The assessment of the Sindhi NLP chatbot involves various strategic scrutinizing ways such
as user surveys, usability test, and qualitative analysis of user's engagement. Accessibility to
SALU students, faculty, and staff members who use the chatbot to clarify their queries, gets
accurate information, and even increase engagement in Sindhi language conversions within
the university community which data also serves as a critical measure of the chatbot's
effectiveness.

Dedicating a Sindhi NLP chatbot to SALU is one of the most important steps to facilitate the
progress in the Sindhi language conservation, digital number skills development and the
transferal of knowledge amongst the agriculture research and academic community. The
application of NLP in SALU's chatbot is innovative in multivarious aspects ranging from
advancing linguistic research, technological innovation, and agricultural education to

22
enabling the users to interact with the platform more efficiently for communication,
collaboration, and learning in Sindhi language.

23
 CHAPTER 5 CONCLUSION, RECOMMENDATIONS AND
LIMITATIONS

The finding and recommendation along the limitation is beauty of any research. Thus, the
recommendation, discussion and conclusion are given in this chapter for looking rapid eye
towards the entire dissertation.

5.1 Conclusion

This study involves a discussion of research design along with stages of creating styles of
feedback that have enduring effects on a student’s later intrinsic motivation in high school.
The data would be obtained from the students and teachers of eight public schools from
district Hyderabad and four of girls' public schools and four of boys. On the other hand, the
study employs a method which is a mix of different methods. For the survey, two types of
questions will be developed in: quantitative and qualitative categories. These findings will
not only help to curate the curriculum for the students but they will also aid the teachers
during students’ beneficial learning.

In this paper we have studied the issues of creation and delivery of Sindhi Language Chatbot
for the SALU (Sindh Agriculture University) people under the umbrella of natural language
processing (NLP). With careful planning, extensive practice, and comprehensive assessment,
it has been possible to testify with validity and effectiveness of the use of NLP technologies
in two ways: for better communication, information retrieval and to enhance engagement in
Sindhi and in the educative domain both.

The Sindhi NLP chatbot constitutes a step in the evolution of the linguistic technology; it is
one of a kind, which widens the communication as well as knowledge retrieval sphere in
Sindhi. Incorporating sophisticated machine learning algorithms together with deep learning
techniques, the chatbot is experienced in terms of comprehension of user queries, reply
generation, and continuous dialogues in Sindhi pertaining various disciplines which match
SALU's academic horizon.

24
Development of given tool was done through coordinative work where effectiveness of it was
mainly built based on collaboration between Linguistics, Computer Science and Agricultural
Sciences. Via a step-by-step optimization and afters-reaction of every step of its performance,
the chatbot's settings reached the niche needs and preferences of SALU users, ensuring
correctness, conversationary value, and linguistic precision of every interaction.

The assessment of the Sindhi chatbot as part of the SALU network has demonstrated the role
that it plays in terms of evaluation and applicability. Through users’ responses, usability
testing and the qualitative feedback analysis highlighting the chatbot as a tool from where the
information is found, assistance is sought, and engagement is felt in the use of Sindhi
language by among students, faculty and staff members is significant.

At the end, the Sindhi NLP chatbot ensures that its provide for further growth, integrating of
extra functionalities, reinforcement of language models and solving of the problems. With
technology developing by the day, the chatbot can be seen as a small stool on the stepping
ladder that introduces digital literacy, preserves the language and raises academic standard for
the SALU community and beyond.

The creation and deployment of the chatbot pertaining to Sindhi NLP at SALU counsel the
language that can be with the technical linguistic skills and helps improving the education,
research and communication in the same language. While we adopt the NPL to the
advantages it can offer, we still keep solemnly adhering to values of the linguistic diversity,
digital inclusivity and knowledge-exchanged by the community among the pursuit of the
academic excellence and the societal impact.

5.2 Limitations

Despite the comprehensive approach taken in this research, it is important to acknowledge

several limitations that may have impacted the findings and outcomes:Despite the
comprehensive approach taken in this research, it is important to acknowledge several
limitations that may have impacted the findings and outcomes:

Data Availability and Quality: The compiling and quality of data from Khairpur University,
majorly being limited to specific semantics with regards to SALU, was a challenge during the
construction of the NLP model. The deficit of the dataset with the strong representation of

25
various audiences could be the reason why the chatbot remain less cogent or work coherently
in some situations.

5.3 Future Work:

The future of this research is bright because it has grown based on derived lessons from the
earlier research, and issues that need attention in the next phase of development are
identified.

Fine-Tuning for Domain-Specific Context: Humanize and fine-tune the conversational

content as well as the language models for Salus in order to have it grasp and generate
domain specific responses in different contexts. Input domain-specific terms, parlance, and
dialectical conventions in the course of achieving the utmost accuracy in the language of
study.

Integration of Multimodal Inputs: Try bringing in many types of information such as text,
graphics and voice to enrich the experience of the users and make the chatting with the
chatbot more accurate. Not only the multimodal NLP techniques will be used, but also their
most advanced version will be used to provide the chatbot with the opportunity to accept and
respond to more types of inputs.

Personalization and Adaptive Learning: Construct personalized learning strategies and

responsive computers algorithms to accommodate the fact that different users have different
styles of learning and be context dependent. Design feedback mechanisms and include
reinforcement learning methods combined to evolve the virtual assistant gradually, boosting
its effectiveness.

Expansion to New Platforms and Channels: Include the chatbot deployment in a range of
expanded channels such as phone applications, social media, as well as voice assistants that
are readily available. The chatbot being an online interaction channel, has a capability to
reach more users all over the globe. It, in addition, simplifies communication and information
retrieval processes.

26
Ethical and Responsible AI Development: Make an ethical accounting by the creation and
deployment of the chatbot which comprise of data privacy, transparency, fairness, and
responsibility. Establish stringent rules about data governance, algorithm transparency and
bias mitigation which would ensure the chatbot strictly dictating the ethical standards and
respecting user rights.

Collaboration and Knowledge Sharing: Collaborate with research institutions and

organizations, businesses and community partners, to avail ourselves of the rich expertise and
applications of NLP researches and technology. Create knowledge-based platforms and
mutual approaches to help in interdisciplinary discourse and implementation of well-
established practices in the linguistic technology building process.

With this prospect in mind, we will be focusing on more ways to increase the performance
and effect of the panel on the SALU domain and beyond. One of the main factors in the
successful deployment of NLP is the constant innovation, collaboration, and ethical AI
implementation. These steps can enable the effective use of NLP that will help users, promote
the language diversity, and as a result, facilitate the academic work in Sindhi studies.

5.4 Summary

This chapter covered the summarize view of the research, limitations, future work and
discussed on the development of chatbot. What strategies have been used and what the work
should be to move for completion have been discussed.

27
References

[1] Mubarak, Z. (2022, April 15). Infant mortality rate falls in Pakistan, but urban-rural disparity
remains, "Dawn Newspaper" https://www.dawn.com/news/1685086

[2] Hamza, N. (2022, May 21). Mortality rate falls in Pakistan, but urban-rural disparity remains,
"Dawn Newspaper" https://www.dawn.com/news/1685086

[3] Wadhera, R. K., Shen, C., Gondi, S., Chen, S., Kazi, D. S., & Yeh, R. W. (2021). Cardiovascular
deaths during the COVID-19 pandemic in the United States. Journal of the American College of
Cardiology, 77(2), 159-169.

[4] Yang, H., Yang, L. C., Zhang, R. T., Ling, Y. P., & Ge, Q. G. (2020). Risks factors for death among
COVID-19 patients combined with hypertension, coronary heart disease or diabetes. Beijing Da
Xue Xue Bao Yi Xue Ban, 420-424.

[5] Li, J., Guo, T., Dong, D., Zhang, X., Chen, X., Feng, Y., & Wan, J. (2020). Defining heart disease
risk for death in COVID-19 infection. QJM: An International Journal of Medicine, 113(12), 876-
882.

[6] Bansal, M. (2020). Cardiovascular disease and COVID-19. Diabetes & Metabolic Syndrome:
Clinical Research & Reviews, 14(3), 247-250.

[7] Kim, J., Huh, J., & Park, D. K. (2023). Classical-to-quantum convolutional neural network
transfer learning. Neurocomputing, 555, 126643.

[8] Rajesh, V., & Naik, U. P. (2021, August). Quantum Convolutional Neural Networks (QCNN)
using deep learning for computer vision applications. In 2021 International conference on recent
trends on electronics, information, communication & technology (RTEICT) (pp. 728-734). IEEE.

[9] Rao, G. E., Rajitha, B., Srinivasu, P. N., Ijaz, M. F., & Woźniak, M. (2024). Hybrid framework for
respiratory lung diseases detection based on classical CNN and quantum classifiers from chest X-
rays.
Biomedical Signal Processing and Control, 88, 105567.

[10] Shah, D., Patel, S., & Bharti, S. K. (2020). Heart disease prediction using machine learning
techniques. SN Computer Science, 1(6), 345.

[11] Singh, A., & Kumar, R. (2020, February). Heart disease prediction using machine learning
algorithms. In 2020 international conference on electrical and electronics engineering (ICE3) (pp.
452-457). IEEE.
28
[12] Ramalingam, V. V., Dandapath, A., & Raja, M. K. (2018). Heart disease prediction using
machine learning techniques: a survey. International Journal of Engineering & Technology, 7(2.8),
684-687.

[13] Jindal, H., Agrawal, S., Khera, R., Jain, R., & Nagrath, P. (2021). Heart disease prediction
using machine learning algorithms. In IOP conference series: materials science and engineering (Vol.
1022, No. 1, p. 012072). IOP Publishing.

[14] Mohan, S., Thirumalai, C., & Srivastava, G. (2019). Effective heart disease prediction using
hybrid machine learning techniques. IEEE access, 7, 81542-81554.

[15] Sharma, V., Yadav, S., & Gupta, M. (2020, December). Heart disease prediction using
machine learning techniques. In 2020 2nd international conference on advances in computing,
communication control and networking (ICACCCN) (pp. 177-181). IEEE.

[16] Abdulsalam, G., Meshoul, S., & Shaiba, H. (2023). Explainable heart disease prediction using
ensemble-quantum machine learning approach. Intell. Autom. Soft Comput, 36(1), 761-779.

[17]Chai, C. P. (2020). The importance of data cleaning: Three visualization examples. Chance, 33(1),
4-9.

[18] Babu, S. V., Ramya, P., & Gracewell, J. (2024). Revolutionizing heart disease prediction with
quantum-enhanced machine learning. Scientific Reports, 14(1), 7453.

[19] Heidari, H., & Hellstern, G. (2022). Early heart disease prediction using hybrid quantum
classification. arXiv preprint arXiv:2208.08882.

[20] Enad, H. G., & Mohammed, M. A. (2023). A Review on artificial intelligence and quantum
machine learning for heart disease diagnosis: current techniques, challenges, and issues, recent
developments, and future directions. Fusion Practice and Application, 8, 25.

[21] Maheshwari, D., Ullah, U., Marulanda, P. A. O., Jurado, A. G. O., Gonzalez, I. D., Merodio, J.
M. O., & Garcia-Zapirain, B. (2023). Quantum machine learning applied to electronic healthcare
records for ischemic heart disease classification. Hum.-Cent. Comput. Inf. Sci, 13(06).

[22] Munshi, M., Gupta, R., Jadav, N. K., Polkowski, Z., Tanwar, S., Alqahtani, F., & Said, W.
(2024). Quantum machine learning‐based framework to detect heart failures in Healthcare 4.0.
Software: Practice and Experience, 54(2), 168-185.

[23] Shankar, A., & Sahana, B. C. (2023). Efficient prediction of runway visual range by using a
hybrid CNN-LSTM network architecture for aviation services. Theoretical and Applied
Climatology, 1-18.

29
[24] Shankar, V., Kumar, V., Devagade, U., Karanth, V., & Rohitaksha, K. (2020). Heart disease
prediction using CNN algorithm. SN Computer Science, 1(3), 170.

[25] Dutta, A., Batabyal, T., Basu, M., & Acton, S. T. (2020). An efficient convolutional neural
network for coronary heart disease prediction. Expert Systems with Applications, 159, 113408.

[26] Mehmood, A., Iqbal, M., Mehmood, Z., Irtaza, A., Nawaz, M., Nazir, T., & Masood, M. (2021).
Prediction of heart disease using deep convolutional neural networks. Arabian Journal for
Science and Engineering, 46(4), 3409-3422.

[27] Sharma, A., Pal, T., & Jaiswal, V. (2022). Heart disease prediction using convolutional neural
network. In Cardiovascular and coronary artery imaging (pp. 245-272). Academic Press.

[28] Sun, Y., Xue, B., Zhang, M., & Yen, G. G. (2019). Completely automated CNN architecture
design based on blocks. IEEE transactions on neural networks and learning systems, 31(4),
12421254.

[29] Bhatt, D., Patel, C., Talsania, H., Patel, J., Vaghela, R., Pandya, S., ... & Ghayvat, H. (2021).
CNN variants for computer vision: History, architecture, application, challenges and future scope.
Electronics, 10(20), 2470.

[30] Xia, K., Huang, J., & Wang, H. (2020). LSTM-CNN architecture for human activity recognition.
IEEE Access, 8, 56855-56866.

[31] Greeks, (03 Feb 2022) Convolutional Block for Image Recognition.

[32] Raza, S., Gul, N., Khattak, H. A., Rehan, A., Farid, M. I., Kamal, A., & Ullah, A. (2024). Brain
tumor detection and classification using deep feature fusion and stacking concepts. Journal of
population therapeutics and clinical pharmacology, 31(1), 1339-1356.

[33] Zhao, X., Wang, L., Zhang, Y., Han, X., Deveci, M., & Parmar, M. (2024). A review of
convolutional neural networks in computer vision. Artificial Intelligence Review, 57(4), 1-43.

[34] Schuld, M., Sinayskiy, I., & Petruccione, F. (2015). An introduction to quantum machine
learning. Contemporary Physics, 56(2), 172-185.

[35] Zhang, Y., & Ni, Q. (2020). Recent advances in quantum machine learning. Quantum
Engineering, 2(1), e34.

[36] Batra, K., Zorn, K. M., Foil, D. H., Minerali, E., Gawriljuk, V. O., Lane, T. R., & Ekins, S.
(2021). Quantum machine learning algorithms for drug discovery applications. Journal of
chemical information and modeling, 61(6), 2641-2647.

30
[37] Ciliberto, C., Herbster, M., Ialongo, A. D., Pontil, M., Rocchetto, A., Severini, S., & Wossnig, L.
(2018). Quantum machine learning: a classical perspective. Proceedings of the Royal Society A:
Mathematical, Physical and Engineering Sciences, 474(2209), 20170551.

[38] Biamonte, J., Wittek, P., Pancotti, N., Rebentrost, P., Wiebe, N., & Lloyd, S. (2017). Quantum
machine learning. Nature, 549(7671), 195-202.

[39] Amin, J., Sharif, M., Gul, N., Kadry, S., & Chakraborty, C. (2022). Quantum machine learning
architecture for COVID-19 classification based on synthetic data generation using conditional
adversarial neural network. Cognitive computation, 14(5), 1677-1688.

[40] Umer, M. J., Amin, J., Sharif, M., Anjum, M. A., Azam, F., & Shah, J. H. (2022). An integrated
framework for COVID‐19 classification based on classical and quantum transfer learning from a
chest radiograph. Concurrency and Computation: Practice and Experience, 34(20), e6434.

[41] Ullah, U., Maheshwari, D., Gloyna, H. H., & Garcia-Zapirain, B. (2022, November). Severity
Classification of COVID-19 Patients Data using Quantum Machine Learning Approaches. In
2022 International Conference on Electrical, Computer, Communications and Mechatronics
Engineering (ICECCME) (pp. 1-6). IEEE.

[42] Ragab, M., Alshehri, S., Alhakamy, N. A., Alsaggaf, W., Alhadrami, H. A., & Alyami, J. (2022).
Machine learning with quantum seagull optimization model for COVID-19 chest X-ray image
classification. Journal of Healthcare Engineering, 2022.

[43] Amin, J., Sharif, M., Gul, N., Kadry, S., & Chakraborty, C. (2022). Quantum machine learning
architecture for COVID-19 classification based on synthetic data generation using conditional
adversarial neural network. Cognitive computation, 14(5), 1677-1688.

[44] Ahmad, G. N., Fatima, H., Ullah, S., & Saidi, A. S. (2022). Efficient medical diagnosis of human
heart diseases using machine learning techniques with and without GridSearchCV. IEEE Access,
10, 80151-80173.

[45] Lakshmanarao, A., Srisaila, A., & Kiran, T. S. R. (2021, February). Heart disease prediction
using feature selection and ensemble learning techniques. In 2021 Third International
Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV)
(pp. 994-998). IEEE.

[46] Noh, Y. D., & Cho, K. C. (2022). Heart Disease Prediction Using Decision Tree With Kaggle
Dataset. Journal of The Korea Society of Computer and Information, 27(5), 21-28.

31
[47] Shukur, B. S., & Mijwil, M. M. (2023). Involving machine learning techniques in heart disease
diagnosis: a performance analysis. International Journal of Electrical and Computer Engineering,
13(2), 2177.

[48] Ahamad, G. N., Shafiullah, Fatima, H., Imdadullah, Zakariya, S. M., Abbas, M., ... & Usman, M.
(2023). Influence of optimal hyperparameters on the performance of machine learning
algorithms for predicting heart disease. Processes, 11(3), 734.

[49] Khandare, A., Agarwal, N., Bodhankar, A., Kulkarni, A., & Mane, I. (2023). Analysis of python
libraries for artificial intelligence. In Intelligent Computing and Networking: Proceedings of IC-
ICN 2022 (pp. 157-177). Singapore: Springer Nature Singapore.

[50] Yalçın, O. G., & Yalçın, O. G. (2021). Complementary Libraries to TensorFlow 2. x. Applied
Neural Networks with TensorFlow 2: API Oriented Deep Learning with Python, 81-94.

[51] Santoso, H. A., Sutoyo, H., Febrian, D., & Putra, Y. S. (2018). Dinus Intelligent Assistance
(DINA) chatbot for university admission services. In 2018 International Seminar on Application
for Technology of Information and Communication (Pp. 417-423). IEEE.

[52] Patel, N. P., Parikh, D. R., Patel, D. A., & Patel, R. R. (2019). AI and web-based human-like
interactive university chatbot (UNIBOT). In 2019 3rd International Conference on Electronics,
Communication and Aerospace Technology (ICECA) (pp. 148-150). IEEE.

[53] Barus, S. P., & Surijati, E. (2022). Chatbot with dialogflow for FAQ services in Matana
university library. International Journal of Informatics and Computation, 3(2), 51-62.

[54] Pi, S. N. M. S., & Majid, M. A. (2020). Components of smart chatbot academic model for a
university website. In 2020 Emerging Technology in Computing, Communication and
Electronics.

[55] El Hefny, W., Mansy, Y., Abdallah, M., & Abdennadher, S. (2021). Jooka: A bilingual chatbot for
university admission. In Trends and Applications in Information Systems and Technologies:
Volume 3 (pp. 671-681). Springer International Publishing;.

[56] Liu, F., Yang, X., & Wang, Y. (2022). An interactive chatbot for university open day. In 2022
IEEE 13th International Conference on Software Engineering and Service Science (ICSESS) (pp.
214217). IEEE.

[57] Gomathy, D. C., & Geetha, D. V. (2021). Voice based university information chatbot system.
International Research Journal of Engineering and Technology (IRJET), 8(7), 1096-1100.

32
[58] Windiatmoko, Y., Rahmadi, R., & Hidayatullah, A. F. (2021). Developing Facebook chatbot
based on deep learning using Rasa framework for university enquiries. In IOP Conference
Series: Materials Science and Engineering, 1077(2).

[59] Le Hoanh Su, T. D. H., Thi-Yen-Linh, T., Thi-Duyen-Ngoc, N., Bao-Tuyen, L., & Ha-
PhuongTruc, N. (2020). Development of an AI chatbot to support admissions and career
guidance for universities. International Journal of Emerging Multidisciplinary Research, 4(2),
11-17.

[60] Mnasri, M. (2019). Recent advances in conversational NLP: Towards the standardization of
chatbot building.

[61] Locke, S., et al. (2021). Natural language processing in medicine: A review. Trends in
Anaesthesia and Critical Care, 38, 4-9.

[62] Shah, S. M. A., Ismaili, I. A., Bhatti, Z., & Waqas, A. (2018). Designing XML tag based Sindhi
language corpus. In 2018 International Conference on Computing, Mathematics and
Engineering Technologies (iCoMET) (pp. 1-5).

33