Design and Implementation of a Heterogeneous Distributed Database System for Students Absence Automation Using Portable Devices

International Journal of Computing and Digital Systems
ISSN (2210-142X)
Int. J. Com. Dig. Sys. 15, No.1 (Mar-24)
http://dx.doi.org/10.12785/ijcds/150175
Design and Implementation of a Heterogeneous Distributed
Database System for Students Absence Automation Using
Portable Devices
Ahmed A. Alsamman1
1
College of Computer Science and Mathematics, University of Mosul, Mosul, Iraq
Received 5 Apr. 2023, Revised 26 Jan. 2024, Accepted 7 Feb. 2024, Published 1 Mar. 2024
Abstract: Students’ absence recording is a fundamental daily process that occurs within colleges, often relying on paper forms, which
is inefficient when the college administration wants to gather absences from multiple remote departments (sites) and calculate the
college’s students’ absence rates. In response to these challenges, a heterogeneous distributed database system was designed to declare
and automate student absences in the colleges using smartphones or laptops and send absences in real-time from inside lecture halls
or laboratories in departments to the college administration to generate absence reports. The system employed a client-server model to
deliver services to the college and its various departments through the local area network (LAN) and the Internet. MSSQL database
tables were designed and fragmented across college sites using horizontal fragmentation technology and then encrypted with the
AES265 algorithm. The proposed system used private SQLite or MSSQL databases for the lecture’s Android or Windows devices. On
the other hand, college administration staff used a secure website built with ASP.NET to access the MySQL database server, which
contains the absence table populated from all sites. The system was protected against false absence declarations by utilizing the local
server’s timing factor, allowing the lecturer only to declare absences on specific days and times. The experimental results showed
high reliability and accuracy of the system and high satisfaction among the users about achieving the desired goals. The findings
demonstrated a strong preference for the system among the lecturers over other systems that relied on student, biometric, and AI-based
devices. At the same time, the college administration staff preferred automatic absence collection using the distributed design of the
system over manual collection.
Keywords: Distributed Database, Heterogeneous Database, Smartphones, Portable Devices
1. Introduction
Students’ absence reporting is a daily routine in educa-
tional establishments as it is a part of the student evaluation
process. Traditional paper-based systems are still in use
worldwide. Still, they are not practical in many respects,
as they waste paper, require time to write names on paper,
and sometimes the lecturers forget to send the absence
sheets to the administration after the lecture, making the
calculation of absence reports inaccurate. Moreover, these
paper sheets will waste time again to read and transfer onto
computers by administrative staff, with possible additional
problems in understanding some of the handwritten names
[1] [2]. As the demand for dependable, scalable, and easy-
access information continues to rise, there is a growing
need for client-server applications and distributed databases
in the modern business environment. Because their data
is spread out across multiple network sites, distributed
database systems facilitate communication and data pro-
cessing improvements. These advantages can be attributed
to the decentralized nature of the system. Not only is data
access quicker, but there is also a decreased likelihood of a
single point of failure, and users have greater local control
over the data [3]. The distributed database (DDB), in turn,
can be described as a set of logically linked databases with
each other and distributed over several sites via computer
networks, as depicted in Figure 1 [4]. As a result of the
development of corporate information and e-government,
the system database of each department constitutes an actual
heterogeneous database framework with its independence
and autonomy inside the network system of many different
functional departments [5]. In this research paper, a dis-
tributed system to report student absences will be designed
and implemented in a college with multiple departments
(sites) using smartphones, laptops, personal computers, and
distributed heterogeneous databases to ensure data inde-
pendence and security. The system is meant to help three
groups: teaching staff, site administration staff, and college
administration staff. It will also make it easier for teachers to
declare absences and make their private reports of absences.
Additionally, departmental administration staff will generate
E-mail address: ahmedalialsamman@uomosul.edu.iq https://journal.uob.edu.bh/

1064 Ahmed A. Alsamman, et al.: Design and Implementation of a Heterogeneous Distributed Database..
Figure 1. Distributed Database Environment
site absence reports. Finally, college administration staff
will use it to obtain absence reports and percentage of
absences. This paper will contribute to building a secure and
reliable distributed system, working as a single unit that fits
a distributed work environment, preserving data dependence
in each department, and prioritizing the validity of recorded
absences and the accuracy of absence reports.
2. Related Work
A lot of work has been dedicated to the problem of
attendance documentation in educational facilities. Some
of these works were based on human biometric features,
utilizing the hardware capabilities of either smartphones
or standalone devices. The hardware includes fingerprint
readers, iris scanners, cameras, and microphones. Atten-
dance documentation systems also use non-biometric de-
vices like RFID readers, iBeacon, and barcode scanners.
Other works have relied on QR codes, the Internet of
Things (IoT), or technologies like Wi-Fi and Bluetooth to
communicate between the lecturer and students. A paper
developed an automated attendance recording system [6]
that stores system data and attendance reports on Android
devices and a MySQL web database server. Attendance
check-in requires students to install the app on their phones.
The system stores data in an Amazon Web Service (AWS)
database. PHP was used to create an administration website.
The system facilitates student attendance management but
requires a smartphone with each student and is prone to
cheating. A Bluetooth-connected smartphone and iBeacon
attendance notification system were created in [7]. The
administration managed student, subject, and staff data on
a website. Students might use their cellphones to enter their
class schedules through Bluetooth to submit attendance data
to the web server, which would alert the lecturer. The system
efficiently monitors student attendance but is also prone to
cheating and requires iBeacon hardware. lecturers and ad-
ministration Using RFID card readers, Arduino UNO, and a
web-based interface [8], researchers designed an attendance
management system for students and staff based on IoT. The
system reads student or employee RFID cards, timestamps
them, and sends them to a web server with a database server
to record attendance and departure following authentication.
Although the system was easy to use and synched with
a real clock, it can be cheated in addition to the cost
associated with Arduino Uno and RFID reader. Researchers
in [2] created an east on-tap system that employed Wi-Fi
peer-to-peer technology to interact with all students. The
instructor’s phone recorded and secured attendance. Once
the lecturer collects attendance, students access the Wi-Fi
and use their app to connect to the instructor’s device to
register their attendance (registration ID). The instructor’s
device discovers neighboring devices, collects registration
IDs using the peer-to-peer connection, and stores them in
their database before sending the attendance record to an
administrative server. Although it is handy, it can be cheated
and requires many smartphones. Barcodes are still active
in attendance systems. A barcode-based student attendance
tracking [9] is developed using various software tools,
including Zint, Scan-IT to Office Program, and Microsoft
Excel. The student’s ID barcode is scanned to record
attendance in the spreadsheet. The system eliminates data
entry errors and improves efficiency but requires a barcode
reader and can be cheated. In the same context, another
system also relies on barcodes [10] that consist of three
parts: an ID cards’ barcode scanner using a web camera
for attendance, an actions recorder for lecturers, and an
administrative activities manager for academic staff. The
system uniquely generates ID cards and lecturer accounts
and sends notifications to parents via SMS or email. The
software uses JavaFX for an adaptive and easy-to-use
application. The system requires the installation of a web
camera and is also prone to cheating. As QR codes became
popular, many QR code-based attendance systems were
created [11][12][13]. The system in [11] uses a geolocation
feature and relies on a web server and MySQL database
to process Android-powered student device requests. GPS
and camera permissions are required to install the student’s
app. After identifying the student’s and device numbers, the
system verifies the student’s location before scanning the
QR code and confirming the student’s attendance. In [12],
a Visual Basic.NET application and QR code scanner are
used to record attendance. Student’s ID cards’ QR codes are
scanned upon entering the room, and the system validates
the log before saving it to MySQL. Classroom reports can
be generated by administrators using the administration
module. The system was highly acceptable among students
but required a QR code scanner and students’ smartphones.
The authors of [13] created a Python program that generates
a unique 21x21 QR code, which is then distributed to each
student via the Internet. Using the PyZBar Python library,
they created a QR code scanner interface to scan and decode
student attendance tracking codes. The system reduced the
student attendance recording but required students to have
smartphones. The proposed system in [14] used Google API
for speech recognition. It records attendance by taking voice
input through an app developed with Kivy and comparing it
to voice samples stored in a database with a 95% accuracy
rate. Although it is cost-effective and highly accurate, it
is not 100 percent perfect. In [15], fingerprint recognition
monitors student attendance. Integrated together are the
Raspberry Pi, fingerprint sensor, and USB converter. After
https://journal.uob.edu.bh/

Int. J. Com. Dig. Sys. 15, No.1, 1063-1074 (Mar-24) 1065
scanning the finger, the database is used to identify the
fingerprint. Reports will reflect the final attendance. The
proposed system stores real-time data and displays it online
for parents and other academic purposes. However, it is
costly concerning the Arduino Uno, Raspberry Pi, and
fingerprint scanner requirements and susceptible to errors.
Another fingerprint-based attendance system [16], com-
prised of an ESP32 microcontroller, a fingerprint scanner,
a keypad, and a buzzer, has been developed. The device
operates in three modes: fingerprint registration, attendance,
and class data input. The easy system achieved good results
but was associated with cost. Several articles have designed
face recognition-based attendance tracking systems utilizing
deep learning and computer vision [17][18][19][20]. In
[17], a Raspberry Pi, camera, and SD card with Raspbian
were used to create a face recognition-based attendance
and door control system. The Raspberry Pi receives images
from the camera and passes them to a pretrained Python
face recognition algorithm (LBPs) that opens the door and
records the student’s attendance in a MySQL database
connected to a PHP website on a web server. Attendance
results were reachable online by any web browser, which
is a good feature, but using Raspberry Pi and a camera
is costly. The proposed system in [18] includes creating
a library of images of students as a first stage, followed
by the training stage using the CNN library, and then
using the camera to take real-time images of the students.
Finally, the images are passed to the Matlab application,
which recognizes the images using the CNN and R-CNN
detectors. The system achieved good results with little
cost but is prone to errors. In [19], a large classroom’s
attendance system stores student photos in a database to
match classroom faces based on cosine similarity. Several
facial photos in the database have enhanced the system’s
ability to differentiate between hazy and distant faces. The
proposed system is fast enough for large rooms, with good
results, but is not error-free. The author of [20] developed an
attendance system using PyCharm for coding, QT Designer
for interface design, MySQL for database management, and
OpenCV face recognition algorithm. The system follows
two stages, face recognition, and attendance recording,
to fulfill attendance system functional requirements with
91% accuracy and real-time performance. It includes face
detection, recognition, information entry, and attendance
data checking, enhancing its structure and function. The
system performance was acceptable but is prone to errors.
An iris recognition-based hybrid system [21] employed
deep learning algorithms and artificial neural networks to
record the attendance of masked persons. The system reads
the iris through a robotic arm equipped with an iris reader
that can automatically and accurately align itself with the
iris. It achieved an accuracy of 96.05 percent. This system
is less effort and highly accurate but not error-free.
In the literature review, it was found that systems rely-
ing on student’s ID cards or students’ reactions through
apps in their smartphones using Bluetooth or Wi-Fi to
submit attendance directly or to scan barcodes or QR
codes techniques are accurate in the process but practically
inaccurate due to the system’s susceptibility to cheating
as students can masquerade the identity of others using
their cards and phones to record illegitimate attendance.
Also, these types of systems are costly and require each
student to have a smartphone, which is not affordable for
some. The systems based on standalone biometric devices
like fingerprint or iris scanners can be expensive and not
necessarily accurate due to recognition errors. The bio-
metric systems utilizing smartphone hardware, like cam-
eras, microphones, fingerprint scanners, and iris scanners,
lack sufficient accuracy and may not even work on entry-
level or cheap devices. Moreover, related works have not
considered data distribution and security, so many sites
(departments) cannot collaborate to serve a higher-level site.
Compared to all previous studies, the proposed system in
this research paper differs in several respects. First, the
system uses distributed databases spread horizontally across
several sites and encrypted with the AES256 algorithm,
resulting in more security and fewer single-point failures.
Second, the system can generate absence reports from a
remote location via the Internet in a distributed manner.
Finally, the system distinguishes itself by not depending on
critical equipment that is either costly or cheatable, such
as fingerprint scanners, iris scanners, personal identification
cards, and students’ phones, to assure the authenticity and
correctness of the data. Therefore, the proposed system
allows all departments to work together within one system
while preserving the security and independence of data in
each department and allowing the college to obtain students’
absences automatically. On the other hand, it prioritizes the
validity of absences and the accuracy of absence reports.
3. System Design Methodology
The proposed system’s implementation depended on the
correct analysis and design steps to ensure it worked well
and kept maintenance costs to a minimum. This is because
flaws or mistakes in the analysis process or logical design
would show up in the final system with an actual payload
and use.
A. Problem Statement
The traditional paper-based systems that colleges use to
keep track of absences have some flaws. For example, some
lecturers are slow to hand out absence sheets, so absence
reports are often wrong. Paper systems are also slower than
electronic ones. Furthermore, the college administration
is disconnected from its sites and wastes time gathering
manual data to create absence reports.
B. Requirements Analysis
Before starting the implementation of the system, some
basic requirements must be analyzed, including:
1) Hardware
• A smartphone with at least Android operating system
version 4.3.x (API 18).
• A server PC running the Windows operating system.

• A network device (a router) that can reserve a static
IP for allocation to the server.
• A hosting website with a MySQL database server.
• A reliable Internet connection.
2) Software
• Microsoft SQL Server.
• Microsoft Visual Studio.
• Android Studio or any Android Integrated Develop-
ment Environment (IDE).
3) Functional Analysis
The system involved three types of users (actors) asso-
ciated with certain functionalities, as follows:
• Lecturer: A lecturer at a site who can login, send
absences to local administration, and issue a class
absence report.
• Head: The head of the department can login, read
data, and issue a site absence report.
• Site Employee: A site administrative employee who
can login, enter data, issue a site absence report, and
send absences to the college administration.
• College Employee: A college administrative em-
ployee who can login, and generate college absence
reports.
• Dean: The college dean who can login, browse ab-
sence totals, and generate college absence reports
• Admin: A college administrative employee who can
manage users and perform database backup and re-
store.
Figure 2 shows the developed system’s user case diagram
(UCD).
C. System Architecture
The system is based on the two-tier client-server model
and employs heterogeneous distributed databases that are:
an MSSQL database that is distributed horizontally over all
sites; SQLite databases at each site; and a remote MySQL
database server used by the college administration, as shown
in Figure 3. The system consists of two main parts:
1) College Departments (sites)
It is a local area network (LAN) comprising a router,
client devices, and an MSSQL database server. It relies on
a two-tier client-server architecture:
• Client: A client can be either a lecturer or a site
administration device.
Figure 2. Use Case Diagram of the System
◦ Lecturer devices: This can be a smartphone
(mobile handset) that works under the Android
operating system with a local SQLite database
server or a PC running a Windows operating
system with a local MSSQL database server.
Clients use a dedicated application to access
data and a local database to keep track of
absence reports related to their subjects. These
devices are connected to the site’s MSSQL
server via Wi-Fi and utilize the server’s time
to keep synchronized.
◦ Site administration devices: a Windows PC run-
ning a Windows application connected to a local
server.
• Server: a Windows PC equipped with an MSSQL
database server. This server’s database is a data frag-
ment with the same structure as other data fragments
but with different data.
2) College Administration
It also relies on a two-tier client-server architecture and
includes:
• Client (administrator): a PC running a website (a web
client) to view college absences and issue absence
reports, relying on the Internet.
• Server: a remote MySQL database server that enables
administrative staff to read the data and issue the
required absence reports.
The ”ADMIN” privileged users in both departments and
college administration interact with the system through a

Figure 3. System Architecture
GUI to manage users, backup, and restore the database.
D. System’s Process Flow Diagram
This section illustrates the process flow at each sub-
system: Windows-based client application, Android-based
client application, and web-based client application.
1) Windows-based application Process Flow
A multi-user application designed for departments to
serve four types of users: the lecturers, employees, the
head of the department, and the system administrator. Four
privileges, ”LEC”, ”EMP, ”HEAD”, and ”ADMIN” are
assigned to the lecturers, employees, the head of the depart-
ment, and the system administrator, respectively. During the
lecture (or lab), the lecturer opens the application from his
smartphone, computer, or laptop, connected to a local area
network (LAN) within the department. When logging in,
users are first authenticated based on credentials (username
and password), and secondly, the privileges are checked up
to permit the user to perform specific actions. Graphical user
interface (GUI) will be customized based on user privilege.
”LEC” privileged users automatically get the current class
(group of students) and subject’s name (course) based on a
weekday and time. Besides, when trying to use the system
at other times outside of the lecture itself, a message will
appear saying that the lecturer does not have a lecture at that
time. As the lecturer taps on the current class (classes), a list
of student names appears ready to be ticked and sent to the
local server. This lecturer can also obtain absence reports
for their students anytime and anywhere through their local
Figure 4. The Process Flow Diagram of Windows-Based Application
SQLite database. ”EMP” privileged users are responsible
for feeding data to the site’s server, including the names
of students, lecturers, and the subject-lecturer assignments.
Besides, they do other jobs like student migration and the
generation of absence reports through a dedicated GUI to
prepare the system to receive the student absences sent by
the teaching staff (lecturers). They also send absences to the
remote MySQL server. ”HEAD” privileged users can issue
department absence reports as needed. Figure 4 shows the
system workflow of a Windows-based application.
2) Android-Based Application Process Flow
This application is only used by lecturers, so no privi-
leges are assigned to the users. The user of this application
will use their credentials to successfully log into the appli-
cation and perform the same actions as the lecturers using a
Windows-based application. Figure 5 illustrates the process
flow of this subsystem.
3) Web-Based Application Process Flow
The users of this application work in the college admin-
istration and are assigned three types of privileges: ”EMP”,
”DEAN”, and ”ADMIN”. The user with ”EMP” privilege
uses the web application to generate college absence reports.
”DEAN” privileged users can browse college absences
and generate reports from the web application. Figure 6
illustrates the process flow of this subsystem.

Figure 5. The Process Flow Diagram of Andoid-Based Application
E. Logical Design of the Database
Data were collected and analyzed using the traditional
paper-based system and interviews with the department’s
administrative employees. After that, the main entities (log-
ical tables) were identified according to their attributes.
For the local administration’s MSSQL database server, six
entities were identified, which are: ”Student”, ”Absence”,
”Subject”, ”Lecturer”, ”User”, and ”Assignment”, as de-
picted in Fig. 7. Only one entity, ”Absence”, was designed
for the client’s local database (see Fig. 8), while three
entities were designed for the remote (college administra-
Figure 6. The Process Flow Diagram of Web-Based Application
tion) MySQL database, namely ”Absence”, ”StageTotal”,
and ”Users” (see Fig. 9). ”Absence” relation (table) at
the college administration is nothing but tuples that are
submitted from each department administration via the
Internet, as shown in Fig. 10. The same principle is applied
to the ”StageTotal” relation.
F. System Implementation
This phase of the design methodology begins with
installing database management systems (DBMS), creating
an MSSQL database at each site, and creating a MySQL
database on the remote server. After this, physical tables
corresponding to logical tables were created inside each
site’s database and lecturers’ local SQLite databases, with
the administration MySQL database server, using the ap-
propriate SQL language. Finally, the graphic user interfaces
(GUIs) for the mobile and PC applications were built and
linked to the various databases of the system via database
connections, according to the system architecture illustrated
in Fig. 3.
• Graphical User Interface (GUI)
The GUIs have been designed in such a way as
to minimize human error using ComboBox and

Figure 7. Site Administration Database Entities
Figure 8. Lecturer’s Private Database “Absence” Entity
CheckBox controls, which are the only ways in
which lecturers and college administration staff can
interface with the system. Local administration staff
can also use the same controls and the TextBox
control to feed data into the system.
◦ Android Application’s GUI: It has many func-
tional GUI’s (see Figure 11) designed using
Figure 9. College Administration Entities
Figure 10. “Absence” Relation at the College Administration
Figure 11. Lecturer’s Android Application GUI
Android Studio) programmed in Java to send
absences to the local server and get absences
reports anytime.
◦ Site’s Desktop Application’s GUI: This GUI
was designed and programmed using Visual C#
and includes several tabs, namely Students, Lec-
turers, Subject Assignment, Absences, Users,
Subjects, Migration (student migration), Re-
ports, Backup & Restore, About, and Sign Out
(see Figure 12). Upon successful login, specific
interfaces appear according to user privilege
”HEAD”, ”LEC”, and ”ADMIN”. For instance,
the Backup Restore tab will only be visible to
the ”ADMIN” user.
◦ College Administration’s Web Application GUI:
This GUI was designed using ASP.NET and
consists of three pages; Default.aspx, Ab-

Figure 12. Desktop Application Tabs
Figure 13. Students Tab
sence.aspx, and Report.aspx. The login page
(default.aspx) displays the login screen and can
be used by authorized college administration
users to log into the website. In contrast, the
absence page (Absence.aspx), lists college ab-
sences. Finally, the report page (Report.aspx)
lists departmental absences and generates ab-
sence reports for the college.
The security of the local sites has been tweaked in
the architecture and the physical design phase. Each site’s
server was first secured as residing on a local area network,
and secondly, the database was set to use the AES265
encryption algorithm.
G. Monitoring and Tuning
The databases have been fed with test data, and every
part of the system has been thoroughly tested and tuned
several times in terms of both the GUI and physical design
to ensure proper workflow.
4. Results
Three user groups: college administration, department,
and teaching staff, have put the system through its paces
with an actual payload. A lecturer successfully received
student absence percentages; lists of student names and sub-
jects (courses) were automatically obtained, and absences
were also successfully sent to the department’s manage-
ment. Some of the lecturer’s android GUIs are depicted
in Figure 11. The departmental staff was successful in
completing many crucial tasks, including managing students
(Figure 13), allocating subjects to the lecturers (Figure
14), migrating students (Figure 15), and generating absence
reports for the department, as shown in Figure 16. They
could do this thanks to their own GUI, which allowed them
to access absences sent by lecturers from the lecture halls.
The college administration staff members could complete
Figure 14. Subject Assignment Tab
Figure 15. Migration Tab
their tasks using their web application GUI. The information
regarding students’ absences for all or specific departments
is displayed in Figure 17. Also, the distributed report
of absence percentages for the college was obtained, as
depicted in Figure 18.
To evaluate the reliability of the developed system,
three questionnaires, Q1, Q2, and Q3, were created for
the three user groups, namely teaching staff, departmental
administrative staff, and administrative staff in the college’s
administration, and emailed to each participant in the group.
The questionnaires contained various evaluation statements
and utilized a 5-point Likert scale with the following
options: strongly disagree, disagree, neutral, agree, and
strongly agree, to indicate the degree of agreement with
each statement. The responses were statistically analyzed
in terms of percentages using the SPSS program.

Figure 16. Reports Tab
Figure 17. Absence Page
Teaching staff: a sample (N) of thirty lecturers were
assigned a questionnaire (Q1) with a set (S) of seven
statements, as follows:
S1: The system is crash-free.
S2: Everyone can use this system because it is simple to
operate.
S3: The system is more truthful than systems based on
student ID cards or phones.
S4: The system is more accurate than biometric-based and
AI-based systems.
S5: For large halls, I prefer quick, semi-accurate gathering
of absences over accurate, slower systems.
S6: Private students’ absence reports generated by this
system for my courses are accurate.
S7: The system meets the lecturer’s needs regarding student
absences.
Thirty responses were received. Table I lists the statistical
analysis of Q1, and the visualization of this analysis is
depicted in Figure 19.
Figure 18. College Absences Report
TABLE I. Statistical Analysis of the Q1
S
Valid
N
Percent %
SD D N A SA
S1 30 30.0 70.0
S2 30 6.7 40.0 53.3
S3 30 6.7 30.0 63.3
S4 30 6.7 6.7 53.3 33.3
S5 30 23.3 16.7 10.0 36.7 13.3
S6 30 36.7 63.3
S7 30 3.3 40.3 56.7
S: Statement, N: Sample size, SD: Strongly Disagree, D: Disagree,
N: neutral, A: Agree, SA: Strongly Agree

Figure 19. Visualization of Q1 Analysis
Departmental Administrative Staff: A sample (N) of
thirty employees were assigned a questionnaire (Q2) with
a set (S) of five statements, as follows:
operate.
S3: Adopting the system assures high absence delivery
rates and quicker response by the lecturers.
S4: The system generates accurate absence percentage
reports for the department.
S5: The system meets the department’s administration
staff’s needs regarding student absences.
Twenty-eight responses were received. Table II shows the
statistical analysis of Q2 and visualization for this analysis
is shown in Figure 20.
TABLE II. Statistical Analysis of the Q2
S
Valid
N
Percent %
SD D N A SA
S1 28 28.6 71.4
S2 28 3.6 17.9 46.4 32.1
S3 28 7.1 53.6 39.3
S4 28 35.7 64.3
S5 28 60.7 39.3
Administrative staff in the college’s administration:
A sample (N) of thirty employees were assigned a
questionnaire (Q3) with a set (S) of four statements, as
follows:
operate.
S3: The system generates accurate absence percentage
reports for the college.
S4: I prefer to have all departments’ absences collected
automatically by a single system as opposed to manually.
S5: The system meets the college’s administrative staff’s
needs regarding student absences.
Twenty-nine responses were received. Table III depicts the
statistical analysis of Q3 and visualization for this analysis
is provided in Figure 21.
TABLE III. Statistical Analysis of the Q3
S
Valid
N
Percent %
SD D N A SA
S1 29 34.5 65.5
S2 29 10.3 27.6 62.1
S3 29 6.9 24.1 69.0
S4 29 20.7 79.3
S5 29 41.4 58.6
Each of the three questionnaires (Q1, Q2, and Q3) was

subjected to Cronbach’s alpha reliability test, as shown in
Table IV, where CA denotes the test value, and the number
of items (statements) is denoted by N. The test results
demonstrated that the three questionnaires are reliable and
internally consistent.
TABLE IV. Reliability Statistics
Q1 Q2 Q3
CA N CA N CA N
0.778 7 0.716 5 0.731 5
CA: Cronbach’s Alpha (0.0 ≤ CA ≤ 1.0), N: No. of statements
5. Discussion
The Cronbach’s alpha (CA) test results in Table IV
showed that Q1, Q2, and Q3 questionnaires had good reli-
ability and internal consistency, scoring 0.778, 0.716, and
0.731, respectively. Generally, CA values greater than 0.5
are acceptable. The reliability of teaching staff (lecturers)
responses was marginally higher (more reliable) than those
of departmental and administrative staff. The vast majority
of users strongly agreed with the system’s error-free op-
eration as well as coordinated with the system’s usability,
report accuracy, and satisfaction with user needs, according
to the statistical analysis of the three questionnaires in
Tables I&II&III. Although system usability looked very
good, it was demonstrated that 94.3% of lecturers agreed
with system usability versus 78.5% and 89.1% for the
departmental and college administrative staff, respectively,
because, from my point of view, 17.9% neutral responses
related to unfamiliarity with comprehensive systems. In
comparison to systems relied on students’ phones via
Internet [6], Bluetooth [7], Wi-Fi [2], barcode [9][10],
QR code [11][12][13], and RFID cards [8], this one has
been endorsed by 93.3 percent of lecturers in Q1 for its
truthiness, and 86.6 percent of them think it is more ac-
curate than biometric-based systems [15][16], and artificial
intelligence-based systems (deep learning and computer vi-
sion) [17][18][19][20]. However, for large lecture halls, only
40% of lecturers -13.3% strongly agreed- preferred speed
over accuracy, and 50% -23.3% strongly disagreed- would
not forego accuracy. In another situation, 92% of employees
in Q2 agreed that the system would ensure that lecturers
would respond to absences more quickly and thoroughly.
Also, 100 percent of employees in Q2 have certified the
accuracy of absence reports and fulfillment of the work
tasks. It was evident in Q3 that all employees (100 percent)
preferred a single distributed system to automatically gather
absence data from remote departments, which is the relevant
feature of the proposed system. The proposed system began
work after accurate data was fed into the system by the
research sample of employees. The flawless design of the
system contributed to facilitating the data entry process with
a minimum of errors through the use of drop-down menus or
selections. Putting the system under experimental use by the
lecturers and employees after feeding it with data for about
a month was important in effectively testing the system and
highlighting its strengths or weaknesses. Obtaining reliable
results from the questionnaires Q1, Q2, and Q3 required
quite a bit of time, especially since the employees have
job responsibilities on the one hand. On the other hand,
patience in completing the questionnaire without rushing
was emphasized because speed in such matters is disastrous
and unreliable. As a consequence, reliable answers were
obtained from the questionnaires.
6. Conclusions And Future Work
Recording absences is one of the basics in evaluating the
academic careers of university students. Still, most colleges
need help collecting these absences from departments that
follow separate and independent systems. This paper con-
tributed significantly to present a solution to these issues
by building a distributed system suitable for the colleges’
structure, consisting of several geographically separated
departments. The system’s sites operate as a single coherent
unit via distributed database technology, saving all parties
the time and effort of handing over and calculating ab-
sence percentages. The system employs predefined lecture
assignments using a combination of weekday and time
schedules that automatically allow the lecturer to receive
a list of student names during the lecture. It also prevents
absences from being declared outside the lecture’s time and
ensures validity. Testing results by the teaching staff and
all employees demonstrate excellent reliability, usability,
and accuracy. Most teachers prefer systems with credibility
and high accuracy over systems that depend on biometric
devices, student response, and artificial intelligence for
small and medium classrooms. On the other hand, Not
a small percentage of the teachers and staff preferred
using these systems for only large classrooms, giving up
only part of the accuracy and credibility for speed. The
limitations of this study were that it was about a segment
of employees who were not accustomed to information
systems, which required giving them the basics of working
on the system. Some showed reluctance or unwillingness to
use the system at first, but things went normally after they
were trained to use the system. So, training is crucial for
some. In addition, some iOS-based smartphone users have
been excluded from testing the Android app because their
phones were not supported. According to the limitations and
the results, the future direction is to improve the system by
incorporating other platforms, notably iOS. Also, the system
can be further improved by adding biometric or AI-based
techniques that can be activated automatically for large halls
and used when time matters.
7. Acknowledgment
I am incredibly grateful to my colleague Dr. Ghassan
J. Mohammed for his support, the Scientific Committee in
the Department of Computer Science/ College of Computer
Science and Mathematics/ University of Mosul, and the
Deanship of the College for their support. Finally, I would
like to thank the anonymous reviewers for their valuable
comments on the earlier version of this paper, which forced
me to spend more time and effort enhancing it, and everyone
who cooperated with me to complete the research.

References
[1] S. Budi, O. Karnalim, E. D. Handoyo, S. Santoso, H. Toba,
H. Nguyen, and V. Malhotra, “Ibats - image based attendance
system: A low cost solution to record student attendance in a
classroom,” in 2018 IEEE International Symposium on Multimedia
(ISM), (Taichung, Taiwan), pp. 259–266, IEEE, December, 2018.
[2] A. Jaiswal, J. S. Mittapalli, B. R. Dilip, and S. Shridevi, “An-
droid based – one tap attendance system,” in 2020 International
Conference on Emerging Trends in Information Technology and
Engineering (ic-ETITE), (Vellore, India), pp. 1–7, IEEE, February,
2020.
[3] B. M. Abdelhafizr, “Distributed database using sharding database
architecture,” in Asia-Pacific Conference on Computer Science and
Data Engineering (CSDE), (Gold Coast, Australia), pp. 1–17, IEEE,
December, 2020.
[4] M. T. Özsu and P. Valduriez, Principles of Distributed Database
Systems. Cham, Switzerland: Springer International Publishing,
4 ed., 2020.
[5] Y. Zhu and Y. Li, “A data sharing and integration technology for
heterogeneous databases,” International Journal of Circuits, Systems
and Signal Processing, vol. 16, no. 2, pp. 232–238, 2022.
[6] A. F. A. Fatah, R. Mohamad, F. Y. A. Rahman, and N. I. Shuhaimi,
“Student attendance system using an android based mobile ap-
plication,” in 2021 IEEE 11th IEEE Symposium on Computer
Applications Industrial Electronics (ISCAIE), (Penang, Malaysia),
pp. 224–227, IEEE, April, 2021.
[7] M. A. Hidaya and H. M. Simalango, “Students attendance system
and notification of college subject schedule based on classroom us-
ing ibeacon,” in 2018 3rd International Conference on Information
Technology, Information System and Electrical Engineering (ICI-
TISEE), (Yogyakarta, Indonesia), pp. 253–258, IEEE, November,
2018.
[8] U. Koppikar, S. Hiremath, A. Shiralkar, A. Rajoor, and V. P.
Baligar, “Iot based smart attendance monitoring system using rfid,”
in 2019 1st International Conference on Advances in Information
Technology (ICAIT), (Farmingdale, NY, USA), pp. 193–197, July,
2019.
[9] A. E. A. A. Salah Elaskari, Muhammad Imran, “Using barcode to
track student attendance and assets in higher education institutions,”
Procedia Computer Science, vol. 184, pp. 226–233, 2021.
[10] L. M. J. Ahmed Yaseen Mustafa, “Electronic student attendance
management system based on id-card,” Sumer University Journal
for Pure Science, vol. 1, no. 1, pp. 56–66, November, 2022.
[11] T. H. T. Khang Jie Liew, “Qr code-based student attendance system,”
in 2021 2nd Asia Conference on Computers and Communications
(ACCC), (Singapore), pp. 20–24, IEEE, September, 2021.
[12] D. M. Acasamoso, E. C. Avila, and S. Vargas, “Development and ac-
ceptability of a student daily attendance monitoring system,” in 2021
IEEE 13th International Conference on Humanoid, Nanotechnology,
Information Technology, Communication and Control, Environment,
and Management (HNICEM), (Manila, Philippines), pp. 1–4, 2021.
[13] S. V. Sai Harsith Reddy, G. R. Sekhar Raju, N. Jayanth, M. C. Sai,
B. Pandey, G. G., and H. Gohel, “Design of qr based smart student
attendance system,” in 2023 IEEE 2nd International Conference on
AI in Cybersecurity (ICAIC), (Houston, TX, USA), pp. 1–4, 2023.
[14] N. Sandhya, R. Vijaya Saraswathi, P. Preethi, K. Aarti Chowdary,
M. Rishitha, and V. Sai Vaishnavi, “Smart attendance system using
speech recognition,” in 2022 4th International Conference on Smart
Systems and Inventive Technology (ICSSIT), (Tirunelveli, India),
pp. 144–149, January, 2022.
[15] R. Memane, P. Jadhav, J. Patil, S. Mathapati, and A. Pawar,
“Attendance monitoring system using fingerprint authentication,” in
2022 6th International Conference On Computing, Communication,
Control And Automation (ICCUBEA, (Pune, India), pp. 1–6, August,
2022.
[16] Maurizfa and T. Adiono, “Smart attendance recording device based
on fingerprint identification,” in 2021 International Symposium on
Electronics and Smart Devices (ISESD), (Bandung, Indonesia),
pp. 1–4, June, 2021.
[17] O. A. R. Salim, R. F. Olanrewaju, and W. A. Balogun, “Class
attendance management system using face recognition,” in 2018
7th International Conference on Computer and Communication
Engineering (ICCCE), (Kuala Lumpur, Malaysia), pp. 93–98, IEEE,
September, 2018.
[18] P. Asmitha and T. Sunitha, “Student attendance using face recog-
nition technology,” in 2022 International Conference on Intelligent
Controller and Computing for Smart Power (ICICCSP), (Hyderabad,
India), pp. 1–4, July, 2022.
[19] D. Mery, I. Mackenney, and E. Villalobos, “Student attendance
system in crowded classrooms using a smartphone camera,” in
2019 IEEE Winter Conference on Applications of Computer Vision
(WACV), (Waikoloa, HI, USA), pp. 857–866, IEEE, January, 2019.
[20] X. Chen, “Study on student attendance system based on face
recognition,” Journal of Physics: Conference Series, vol. 2492,
no. 1, pp. 1–6, May, 2023.
[21] A. Bhardwaj, A. Mehta, A. M. Gaikwad, V. Vyas, and R. Mahajan,
“Contactless mask monitoring and biometric attendance system,” in
2022 International Conference on Signal and Information Process-
ing (IConSIP), (Pune, India), pp. 1–5, August, 2022.
Ahmed A. Alsamman received an MSc
in Computer Science from the University of
Mosul in 2012. He joined the teaching staff
at the computer science department, College
of Computer Science and Mathematics at
the University of Mosul in 2012 and started
his academic career. He is a PhD student in
the computer science department, College of
Computer Science and Mathematics, Univer-
sity of Mosul. His research interests lie in
information systems, IoT, image processing, and data security.

Design and Implementation of a Heterogeneous Distributed Database System for Students Absence Automation Using Portable Devices

More Related Content

Similar to Design and Implementation of a Heterogeneous Distributed Database System for Students Absence Automation Using Portable Devices (20)

Recently uploaded (20)

Design and Implementation of a Heterogeneous Distributed Database System for Students Absence Automation Using Portable Devices