APPLIED ARTIFICIAL INTELLIGENCE

2022, VOL. 36, NO. 1, e2083796 (1137 pages)

https://doi.org/10.1080/08839514.2022.2083796

Development and Evaluation of an Attendance Tracking

System Using Smartphones with GPS and NFC
Te-Wei Chianga, Cheng-Ying Yangb, Gwo-Jen Chiouc, Frank Yeong-Sung Lina, Yi-
Nan Lind, Victor R. L. Shene,f, Tony Tong-Ying Juange, and Chia-Yang Line
a
Department of Information Management, National Taiwan University, Taipei City, Taiwan, ROC;
b
Department of Computer Science, University of Taipei, Taipei City, Taiwan, ROC; cDepartment of
Electrical Engineering, National Formosa University, Yunlin County, Taiwan, ROC; dDepartment of
Electronic Engineering, Ming Chi University of Technology, New Taipei City, Taiwan, ROC; eDepartment
of Computer Science and Information Engineering, National Taipei University, New Taipei City, Taiwan,
ROC; fDepartment of Information Management, Chaoyang University of Technology, Taichung City,
Taiwan, ROC

ABSTRACT ARTICLE HISTORY

As usual, if many students are attending the lectures, their Received 12 February 2022
attendance tracking may become time consuming. Revised 21 May 2022
Furthermore, there are some possibilities that students could Accepted 25 May 2022
cheat the lecturers of their attendance in the classroom.
Therefore, a reliable method to manage the attendance tracking
has become a critical issue. This paper aims to propose an
attendance tracking system using an Android smartphone
equipped with Global Positioning System (GPS) and Near Field
Communication (NFC) technologies. Lecturers and students can
constantly connect with one another by using smartphones to
check and show their attendance automatically if they down­
load and install the software Application (App). Finally, the
experimental results have shown that our proposed system
can successfully reduce some time for tracking students’ atten­
dance. It also allows users to use their own Android smart­
phones without purchasing other electronic devices.

Introduction
Advances in smartphone hardware and software technologies not only made
the teaching task easier for teachers, but also helped students learn some
subjects more efficiently in and out of class. Nevertheless, student attendance
tracking is still one of the most pressing issues for professors. With the huge
popularity of smartphones, many studies have emerged on comparing those
elements such as the system security and reliability of two most famous mobile
platforms, iOS and Android (Strnadelova, Halamova, and Kanovsky 2019).
One method to solve the problem of tracking attendance on the Android
platform was previously proposed in reference (Noor et al. 2016).

CONTACT Victor R. L. Shen [email protected]; [email protected] Department of Computer

Science and Information Engineering, National Taipei University, 151, University Rd., Sanxia District, New Taipei City
237, Taiwan, ROC
Subject Area: Intelligent System Development Reference No.: 2022-UAAI-0039
© 2022 The Author(s). Published with license by Taylor & Francis Group, LLC.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http://
creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any
medium, provided the original work is properly cited.
e2083796-1118 T.-W. CHIANG ET AL.

Student attendance is an important issue in all universities. Students in

class usually complain that the conventional method of tracking atten­
dance leads to the loss of their rights to be educated due to time
inevitably wasted in the checking process. On the other hand, professors
who need to follow the course syllabus on schedule are also concerned
about more time taken to call each student by name. Furthermore,
professors are aware that students leave class during lecture hours or
find other ways to falsely track their attendance (Mittal et al. 2016).
However, it is still possible for students to be able to fake their attendance,
and the existing approaches are unable to avoid these unusual cases. For
example, some students come to class after ensuring their professors who
have marked their attendance, and then they leave the classroom during
lecture hours without asking for permission. Of course, this unusual case
should be prohibited (Kamaraju and Kumar 2017). Thus, professors need
a reliable and more efficient system to solve this problem and stop the unusual
case. It is unnecessary to call each student’s name from the list continually
throughout the lecture hours, and then professors can manage the student
attendance more efficiently.
Therefore, this paper aims to develop a student attendance tracking system,
which not only makes sure that students come to class by checking their identi­
fication, but also constantly monitors their locations using Global Positioning
System (GPS) and Near Field Communication (NFC). For achieving the goal of
solving the attendance tracking problem, the proposed system is environmentally
friendly because it is paperless (Nguyen and Chew 2017). Because a peer-to-peer
mode is applied, the data exchange works in both directions between two mobile
devices communicating with each other via NFC. Common applications (Apps)
are used to exchange contact information (vCard) between two smartphones.
Hopefully, less equipment is needed to reduce some time taken to track students’
attendance. However, the cheating problem is assumed to be eliminated, for
instance, if a student lends his/her smartphone to a colleague, or if someone else
is using another person’s smartphone during the lecture hours.
The rest of this paper is organized as follows: Section 2 reviews the GPS,
NFC, and the related works. The overview of our proposed system and
methodology are described in Section 3. The experimental results are pre­
sented in Section 4. Section 5 remarks the conclusion with possible research
areas for future system development.

Literature Review
In this Section, some fundamentals regarding global positioning system (GPS),
near field communication (NFC), and the related works are presented.
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1119

Global Positioning System

The GPS technology uses a receiver to get the geolocation and time informa­
tion of devices via satellites. Using the unobstructed line of sight of more than
four GPS satellites, we can obtain devices’ location accurately anywhere on the
earth (Shi, Sim, and Lee 2017). Recently, people can receive GPS information
from smartphones with an advantage in creating the indoor positioning
system. For example, one App, called Travel Portal, relies on GPS and Geo
Tagging to create a system where people can manage their trips in synchro­
nization with travel agencies and answer the travel queries (Fridman et al.
2016). Another App focuses on children’s safety by tracking their location
through GPS on the Android device (Maryam et al. 2016; Sunehra, Priya, and
Bano 2016). Meanwhile, both stand-alone GPS devices and smartphones with
GPS technology work by detecting the position and signals from a series of
satellites set up by the US military. The devices calculate their position by
reference to these satellites (Walkhighlands 2021).
In addition, AGPS (Assisted Global Positioning System) was also used for
indoor positioning. It is important to underline that the indoor environment
has problems with GPS/GNSS (Global Navigation Satellites System) position­
ing. These problems are reduced by some satellites recently, allowing the
visibility of the constellation even in rooms with average windows (Hay,
Duffy, and Whitfield 2017). GNSS positioning would be generally possible
also in concrete buildings. Practically, GNSS signals may be detectable at some
places in a concrete building, but it is generally agreed that the current
smartphones cannot reliably receive and decode GNSS signals inside a multi-
story concrete building.
Furthermore, four satellites are used to have the three-dimensional posi­
tion, if instead the two-dimensional position is enough, three satellites are
enough. Most GPS/GNSS Apps installed on the cellular terminals (for exam­
ple, all road navigation systems) work with two-dimensional positions that
allow positioning even with poor coverage (Faustmann et al. 2020). It
describes a system with selective repeaters, which imposes the need of instal­
ling and adjusting infrastructure. While a mobile phone could not provide
positions more accurate than five meters. This potentially allows one to
accurately verify whether the student is really in the classroom or in the library
next door.

Near Field Communication

The NFC technology is one of the methods being used to exchange datasets in
many scenarios. It is a standard which covers communication protocols and
data exchange (Mohandes 2015). Since our proposed system includes the
implementation of NFC technology on Android devices as one of the
e2083796-1120 T.-W. CHIANG ET AL.

functions to help students confirm their presence. Another related paper about
using NFC on Android devices has been published to reduce working time and
to increase their productivity.
Many papers have focused on the NFC technology for face recognition used
on Android smartphones as a tool for validating identification (Rana and
Ciardulli 2017). The App helps users tag the NFC to immediately validate
the users’ identity when they are in danger or in emergency. It could be used in
many scenarios, e.g., healthcare, public transportation, and neighborhood
surveillance.
Furthermore, the NFC technology has a potential function to be imple­
mented in the attendance tracking system as there is a method of applying
the Android device Nexus7. It needs to have the lowest energy cost among
all tablets, and to let students confirm their attendance with ease (Sousa
and Barata 2020). However, when using the previous methods, students
could easily cheat and leave class without asking for permission, and they
would still have their presence recorded on the students’ attendance sheet.
This study presents the best option to prevent this unusual case from
occurring.

Related Works
There are many methods to check the student attendance, including
fingerprint biometrics (Mittal et al. 2016), high real Zigbee (Kamaraju
and Kumar 2017), RFID (Nguyen and Chew 2017) and face detection
(Rekha and Ramaprasad 2017; Varadharajan et al. 2016). The fingerprint
biometric system uses a fingerprinting device to accurately identify peo­
ple’s identity. Maddu Kamaraju, et al. proposed a method based on high
real Zigbee to save time in recording students’ attendance (Kamaraju and
Kumar 2017).
With the popularity of smartphones in recent years, numerous Apps have
been developed to help people solve a wide range of problems (Anjana,
Nappinnai, and Sahari 2016). Especially, in the educational sector, there are
many proposed systems which use the iOS or Android platform to help
students learn some knowledge from the lectures more effectively (Douglas,
Mazzuchi, and Sarkani 2020; Faustmann et al. 2020; Hay, Duffy, and Whitfield
2017; Koubaa, Boukattaya, and Damek 2018). By using the Android-based
indoor positioning system, one App used the phone camera to scan the
barcodes on students’ ID cards for identifying students (Kamaraju and
Kumar 2017).
Another system was developed on Android which allowed students to use
their mobile devices, or teacher’s terminal devices, registering their attendance
as well as using a BLE (Bluetooth Low Energy) beacon to transmit a magic
number to prevent students from cheating at registration time (Noguchi et al.
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1121

2018). Although these papers present some methods for checking students’
presence, they cannot still avoid the possibility that students might leave class
after registration.

Proposed Attendance Tracking System

In this Section, the details in our system development are explained. We
require GPS- and NFC-enabled Android devices to help students confirm
their presence and allow the professor’s phone to keep monitoring students’
location until the lecture is over. In the App development part, we used
Android SDK tools and Eclipse that fully supported our present system
development. The reason why we selected the Android operating system is
because of the rapid growth of smartphone users. It is reasonable to expect that
all the students in class have their own smartphones, as would the profes­
sors do.

System Requirements
The presented App system aims to help the educational systems in colleges
become more effective and productive. The student attendance tracking sys­
tem uses smartphone devices based on the technological trends over the last
few years. Many people use smartphones, not only in Taiwan but also every­
where in the world. In the universities of Taiwan, professors sometimes ask
students to do research in class by using their smartphones. Expecting each
one to have an Android smartphone, we have designed the methodology based
on the Android device with GPS and NFC. After the system requirements are
specified correctly, the system design is performed, and then the system
implementation is finished.

System Development

This paper presents a student attendance tracking system using the Android
smartphones with GPS and NFC. Users only need to download the App on
their smartphones without purchasing other hardware systems. By referring to
the system architecture, as shown in Figure 1, we can view the system which
consists of Android Apps and the server with a database system. They are both
connected with a software system, one NFC tag, and an Android smartphone
in hardware requirements.
When using this system, students and professors first need to download
and install the App. The users are required to use their staff ID card or
student ID card to sign up the App before they login. The Android phone can
read the information from the chips, and the system can check the users’
identification. Both students and professors can access the information from
e2083796-1122 T.-W. CHIANG ET AL.

Figure 1. System architecture.

the lectures. At this time, students are asked to tag the NFC to register their
attendance in class. The students’ Android devices automatically upload
their data items to the attendance sheet in the database system. After
successfully uploading the students’ present attendance, the students’ phones
continually update their locations in the database system, and the professor’s
phone tracks students’ attendance statuses throughout the lecture hours.
As shown in Figure 2, professors collect the information from the database
system. If the distance between professor and students is less than a preset
value, e.g. 10 meters, based on the size of the classroom, the App displays
students’ status as “Presence.” However, if the distance is larger than 10
meters, the App turns the status into “Absence.” The diagram shows that the
App uses GPS to help the professor monitor students’ locations in this system.
In the monitoring step, the App on the students’ phones keep updating the
location through the GPS. The App also updates the online database system.
The online database system is composed of one table which is used to store
those data items such as the students’ and professor’s username, password, ID
number, course number, course title, and the distance between students and
professor.
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1123

Figure 2. Attendance scenarios.

Database System Development

The App gets data items from the table. It can calculate the distance between
students and professor on the table automatically, including Pid, Student ID,
Longitude, and Latitude Distance. Table 1 shows students’ personal data items,
including student ID, longitude, latitude, create time and update time. These
data items are updated once they finish logging in the App and tagging the
NFC. Create time is not changed because the time is created at the time shown.
As the first column shown, Pid is a primary key. The primary key cannot be the
same as in other rows. It is used to identify and find out every student’s row of
data items selected.
Table 2 shows that professors use the App to get information from the
database system. This App is designed to automatically calculate the distance
between professor’s smartphone and students’ ones. As this paper uses the
Haversine formula (refer to Eq. 1) (Liu et al. 2020), the distance (d) between
two phones is obtained by calculating two longitudes and latitudes which
we’ve gotten, where ϕ1 and ϕ2 denote the latitudes of two points; λ1 and λ2
denote the longitudes of two points; and γ denotes radius of the sphere.

Table 1. Students updating the database using smartphones.

Pid Student ID Longitude Latitude Create time Update time
1 710483666 121.4558988 25.0054492 2020–03-07 2020–03-07
16:26:16 13:26:16
2 400410824 121.4559908 25.0052870 2020–03-07 2020–03-07
16:20:18 13:26:19

Table 2. Professors getting the information from database system.

Pid Student ID Longitude Latitude Distance Status
1 710483666 121.4558988 25.0054492 12.0 m Presence
2 400410824 121.4559908 25.0052870 12.0 m Presence
e2083796-1124 T.-W. CHIANG ET AL.

Finally, the professor obtains the distance. If the distance is larger than 15
meters, the status in the table is updated as “Absence.” If the distance is less
than 15 meters, the status on the table is updated as “Presence.”
sffiffiffiffiffiffiffiffi�
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi�
ffiffiffi!
ϕ ϕ λ 2 λ 1
d ¼ 2r arcsin sin2 2 1
þ cosðϕ1 Þ cosðϕ2 Þ sin2 (1)
2 2

User Navigation Development

The App in the present system plays a critical role. Each flowchart is explained
stepwise and from figure to figure. In the student’s phase, as shown in Figure 3,
students need to download and install the App. When students launch the
App, they are guided to the sign-up/login page. If students want to sign up,
they are supposed to use their student ID. After they sign in successfully, the
App guides them to the course information page. Students can check the
course information in this page and choose the course they want to take.
Once they have chosen the course, the App asks them to confirm their
attendance with the NFC tag. Then the App continues to update the location
for the professor until the lecture is over. There is no limit to the number of
NFC tags. The professor can provide more than one tag to help students
activate the NFC tag quickly, so they do not need to wait in queue. The
more NFC tags the professor provides, the quicker the students can go through
the entire process.
In the professor’s phase, as shown in Figure 4, the professor also needs to
download, install, and launch the App. The professor needs to sign up and
does the same as the students. After logging in via the lecture information
page, the professor can view and choose all the course information. Once the
course is chosen, the professor can view students’ IDs and the distance
between professor and the students in class. The professor’s smartphone can
automatically monitor the students, ensuring that they are all in the classroom
until the lecture is ended.

App System Implementation

Since the software development is important for the proposed system, Eclipse
was used to develop an App system. The operational procedure of the App
system consists of four parts, as shown in Figure 5.
The first part, which is the most important one, is to develop GUI
(Graphical User Interface). It refers to the users’ experiences in using the
App. Once the GUI was developed, the users can easily find the infor­
mation in the App, i.e. how to log in the system, how to get the course
information, and how to guide the users to tag on the smartphones.
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1125

Start

Install
Application
.app

Registration No

Students Login

Login
No
Successful

Show The Class's

Information of
The Lecture

Open NFC and

Start Scanning

Keep Updating
Students’ Location by
Using GPS until Class
Finished

Stop

Figure 3. Flowchart of student navigation.

e2083796-1126 T.-W. CHIANG ET AL.

Figure 4. Flowchart of professor navigation.

Figure 5. Flowchart of system software.

The second part is to create the connection between the Android phone
and the database system. In this part, the App is designed to allow
smartphones to get data items and upload data items into the database
system. For the connection, the App is required to use the SQLite.
SQLite stores information in the online database system, e. g. students’
location and course number, etc. To connect the online database with
SQLite to smartphones, this study has designed eight programs in PHP.
With these programs the Android App can easily connect with the
online database system, update the table, delete the table, and create
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1127

the row of data items on the table. The third part is GPS development.
Although one can easily get the geolocation on the Android smartphone,
this paper establishes a function to calculate the distance between the
phone and others. The fourth part is the NFC function, in which the
App is designed to read the information via the staff ID cards and
student ID cards. Also, students show their attendance based on the
NFC tagging action.

Figure 6. Graphical user interface.

e2083796-1128 T.-W. CHIANG ET AL.

As shown in Figure 6, the Android App was created by using Eclipse (Eclipse
Foundation-Eclipse 2020). The graphical user interface (GUI) is designed to
improve users’ usage experiences. To reduce the complexity of GUI, the App
uses less than three buttons and three edit-texts in every page. Users can easily
find the button, type in their information, and follow the instructions.
To completely develop this App system, 14 projects of source codes and 12
pages of .xml files were used. The content includes connecting the button with
.xml files and projects, connecting the edit-text between .xml files and projects,
directing the users to the next page, saving the information in the next page,
connecting with the online database system, getting the user’s location, creat­
ing a function to identify NFC tags, calculating the distance between professor
and students, and updating the information in the online database system.
This system needs a server to reserve all the App information. The environ­
ment of the server requires Apache, PHP, MySQL and phpMyAdmin. Our
proposed database system is created by using XAMPP (Apache Friends 2020).
phpMyAdmin is used to manage the information in the database system
directly. We use phpMyAdmin to create the name of the database and the
columns of data items including primary key, Pid.
Professors can not only get the students’ location by smartphones, but also
get the information from the database system. Without using the smartphone,
professors can manage the database system in a computer. Our proposed
system uses Notepad++ to write the PHP codes for making the connection
between the App and the online database system (Don Ho-Notepad++, 2020;
Pedro et al. 2019). And the PHP files are also used to manage all the informa­
tion which includes storing, updating, and deleting the columns of data items
in the database system.
In the present system, the HTC new one has been used to help us test the
Android App. It is necessary that users should use a computer to run the online
database system and a smartphone, which supports Android 2.3.3 or higher
versions with the function of NFC, for downloading and running the Android
App with at least one NFC tag which can read or write some information.

Experimental Results and Performance Evaluation

In this Section, all the system design based on the previous ideas is presented.
With only one App, students can easily check the information of the course,
and show their attendance status without spending much time. Professors can
also use the same App to manage students’ attendance without calling stu­
dent’s name. When applying this system, users are only required to have their
own Android devices, and professors can decide how many tags to be
provided.
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1129

Activity Flow on Android Devices

Users are required to choose their position before they are guided to different
pages. In the users’ login page, they are required to type in their username and
password. If users do not register yet, they need to click the “sign-in” button,
which guides users to the sign-in page.
Students can get the class information once they login the system success­
fully. They are asked to click the course number to choose the class that they
are going to attend. Students can see the course number they have chosen
previously. The students are asked to open the NFC and get close to the NFC
tag, confirming his or her attendance in class.
After tagging the NFC successfully, as shown in Figure 7, the smartphone
can show messages about the related information, and ask the students to stay
in class while the location monitoring mode is turned on. Figure 8 shows the
students’ information required by the professor. Then the professor can know
the students’ distance and their attendance status.

Performance Evaluation

In this Sub-section, our proposed system was compared to others. There are
five approaches available for performance evaluation. The first one used GPS
and NFC to monitor the students’ location. The second one used face detec­
tion to identify students’ personal information through the camera (Rekha and
Ramaprasad 2017). The third one used the Android system with a camera to
check the students’ ID card (Varadharajan et al. 2016). The fourth one used the
fingerprint detection machine to identify the students’ fingerprint when they
wanted to show their attendance (Kamaraju and Kumar 2017). The fifth one
used the Bluetooth with a beacon to show students’ attendance (Nguyen and
Chew 2017).
As shown in Table 3, the number of hardware requirements in our pro­
posed system is two, which is the fewer one. And it can keep monitoring
students’ location unlike other approaches, which provides monitoring stu­
dents’ location only one time. The final results have also shown that students
cannot cheat their attendance by using our proposed system.
This paper used the online survey. Users were asked to read the instructions
before answering the questionnaire and to answer all the questions based on
their own experiences. In the experiment, there are 67 sets of questionnaires
sent out and 60 sets of valid questionnaires taken back, including 49 students
and 11 professors (Acknowledgments). In Table 4, we compare all the users’
satisfaction degree, the percentage of users who think it is time consuming, the
order of users’ satisfaction, the average value of users’ execution time, and the
correct rate of users’ experiences in presenting the advantages of our proposed
system.
e2083796-1130 T.-W. CHIANG ET AL.

Figure 7. Notification of tagging successfully.

From the experimental results shown in Table 4, our proposed system has
the highest percentage of users’ satisfaction degree. Based on the users’ usage
experiences, most of users think that the execution time of our proposed
system is less than others. We calculated the average value of users’ execution
time by simulating that there are 50 students in class. The results have shown
the average time of our proposed system only takes 75 seconds and the face
detection takes the longest time to finish. Thus, we can see the consistency of
our approach is better than that of others.
To compare our proposed system to other approaches, the real situation has
been simulated. After calculating the total running time, as shown in Figure 9,
our system with one tag was compared to the face detection. Because “Face
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1131

Figure 8. Students’ information.

Detection” approach is located as the second order of users’ satisfaction

according to the survey results, this is the main reason why only “Face
Detection” approach was selected. The experimental results have shown that
for more students in class, the face detection took much longer time than our
proposed system.
From the experimental results shown in Figure 10, our proposed system
with two tags is two times faster than the one with only one tag, and it is also
much faster than the face detection. Figure 10 shows that the more tags we
provide, the shorter the total running time becomes. Even though our
proposed system only provides one tag, it is still faster than the face
detection.
 e2083796-1132

Table 3. A functional comparison of different attendance management systems.

T.-W. CHIANG ET AL.

Our Proposed Face Detection (Rekha and Android Smartphone Fingerprint Approach (Kamaraju and Bluetooth Approach (Nguyen and
Approaches System Ramaprasad 2017) (Varadharajan et al. 2016) Kumar 2017) Chew 2017)
Metrics (1) (2) (3) (4) (5)
Hardware requirement Smartphone Smartphone, Computer, Camera Smartphone, Computer Fingerprint Recognition Device, Smartphone, Computer
Computer
Other hardware Server Server Server Server Server
requirement
Total number of devices 2 3 2 3 2
needed
Monitoring all the time Yes No No No No
Cheating attendance May not Yes Yes May not Yes
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1133

Table 4. Experimental results of different approaches.

Face Detection Android Fingerprint Bluetooth
Our (Rekha and Smartphone Approach Approach
Proposed Ramaprasad (Varadharajan (Kamaraju and (Nguyen and
Approaches System 2017) et al. 2016) Kumar 2017) Chew 2017)
Metrics (1) (2) (3) (4) (5)
Users’ satisfaction 88.33% 80% 70% 68.33% 35%
degree
The percentage of users 13% 85% 15% 46.67% 43.33%
who think it is time
consuming
The order of users’ 1 2 3 4 5
satisfaction
The average value of 75s 500s 80s 450s 200s
users’ execution time
The correct rate of users’ 93.3% 95.2% 89.5% 92.6% 73.3%
experiences in
presenting the
advantages

In addition, in references (Iio 2016; Kinoshita, Niibori, and Kamada 2019;

Takahashi, Niibori, and Kamada 2021) other related student attendance man­
agement systems were proposed. Iio’s system is based on student’s QR-code
but it has no increase of feedback comments (Iio 2016). Kinoshita et al.’s
system maps participants onto the seat map but it has low interactivity
(Kinoshita, Niibori, and Kamada 2019). Takahashi et al. proposed a web-
based attendance management system with NFC seat identifiers but it is easy
to make cheating (Takahashi, Niibori, and Kamada 2021). In summary, our
proposed system can outperform others.

Figure 9. Total running time of face detection compared to the proposed system with one tag.
e2083796-1134 T.-W. CHIANG ET AL.

Figure 10. Total running time of face detection compared to the proposed system with two tags.

Conclusion
Currently, many studies have been done to address the attendance tracking
issue, yet most methods require the interested users to buy and maintain
additional expensive electronic devices. Thus, because of different methodol­
ogies, those systems are simply uneconomical and unattractive to users. This
paper has successfully developed a promising attendance tracking system
which achieves the goal of low cost and minimal execution time.
The contributions of this paper are presented as follows:

(1) Users only need to run the software App by using their own Android
smartphones with GPS and NFC without purchasing additional elec­
tronic devices and save more money.
(2) Professors can take shorter time to track students’ attendance.
(3) The smartphone can monitor students’ attendance status automatically
and efficiently.

In the future, our work will follow the growing popularity of the latest
fingerprint-identity functionality on smartphones. Both Android and iOS
smartphones can use the fingerprint recording technology to confirm their
identity with the combination of our system to make the attendance tracking
system work seamlessly. Furthermore, the GNSS availability and accuracy also
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1135

need to be carefully investigated and applied for different situations/rooms/

floors/buildings to improve the system performance.

Acknowledgments
The authors are grateful to the anonymous reviewers for their constructive comments, which
have improved the quality of this paper. Also, in the online survey, those 49 students and 11
professors are also highly appreciated for their efforts and time in answering all questions.

Disclosure Statement
No potential conflict of interest was reported by the author(s).

Funding
This work was supported by the Ministry of Science and Technology, Taiwan, ROC, under
grants MOST 107-2221-E-845-001-MY3 and MOST 110-2221-E-845-002-.

References
Anjana, G., S. Nappinnai, and P. Sahari. 2016. Smart home system using a wireless sensor
network for elderly care. Proceedings of Second International Conference on Science
Technology Engineering and Management (ICONSTEM), Chennai, India, 51–55.
Apache Friends. XAMPP. Accessed May 22, 2020. https://www.apachefriends.org/zh_tw/
index.html.
Don Ho-Notepad++. Accessed November 24, 2020. https://notepad-plus-plus.org/zh.
Douglas, A., T. Mazzuchi, and S. Sarkani. 2020. A stakeholder framework for evaluating the
utilities of autonomous behaviors in complex adaptive systems. Systems Engineering
23 (5):100–22.
Eclipse Foundation-Eclipse. Accessed November 7, 2020. https://www.eclipse.org.
Faustmann, C., M. Bajzek, H. Hick, F. Edtmayer, and S. Walch. 2020. System models and
model classification in tribological system development. Systems Engineering 23 (6):783–94.
doi:10.1002/sys.21562.
Fridman, L., S. Weber, R. Greenstadt, and M. Kam. 2016. Active authentication on mobile
devices via stylometry, application usage, web browsing, and GPS location. IEEE Systems
Journal 11 (2):513–21. doi:10.1109/JSYST.2015.2472579.
Hay, L., A. H. B. Duffy, and R. I. Whitfield. 2017. The S-cycle performance matrix: Supporting
comprehensive sustainability performance evaluation of technical systems. Systems
Engineering 20 (1):45–70. doi:10.1002/sys.21378.
Iio, J. 2016. Attendance management system using a mobile device and a web application.
Proceedings. of 19th International Conference on Network-Based Information Systems
(NBiS), Ostrava, Czech Republic, 510–15. doi:10.1109/NBiS.2016.44.
Kamaraju, M., and P. A. Kumar. 2017. Wireless fingerprint attendance management system.
Proceedings. of IEEE International Conference on Electrical, Computer and
Communication Technologies (ICECCT), Coimbatore, Tamil Nadu, India, 1–6.
e2083796-1136 T.-W. CHIANG ET AL.

Kinoshita, S., M. Niibori, and M. Kamada. 2019. An attendance management system capable of
mapping participants onto the seat map. In Advances in Network-Based Information Systems
NBiS 2018 lecture notes on data engineering and communications technologies, ed. L. Barolli,
N. Kryvinska, T. Enokido, and M. Takizawa, 22. Cham: Springer. doi:10.1007/978-3-319-
98530-5_80.
Koubaa, Y., M. Boukattaya, and T. Damek. 2018. Adaptive sliding mode control for trajectory
tracking of nonholonomic mobile robot with uncertain kinematics and dynamics. Applied
Artificial Intelligence 32 (9–10):924–38. doi:10.1080/08839514.2018.1519100.
Liu, B., Y. Yi, H. Shen, and C. Niu. 2020. Composite observer-based integral sliding mode
dynamical tracking control for nonlinear systems subject to actuator faults and mismatched
disturbances. Measurement and Control 53 (7–8):1309–17. doi:10.1177/0020294020923077.
Maryam, A., N. Siddiqui, M. A. Qadeer, and M. S. Umar. 2016. Travel management system
using GPS & Geo tagging on android platform. Proceedingss of Thirteenth International
Conference on Wireless and Optical Communications Networks (WOCN), Hyderabad,
India, 1–6.
Mittal, Y., A. Varshney, P. Aggarwal, K. Matani, and V. K. Mittal. 2016. Fingerprint biometric
based access control and classroom attendance management system. Proceedings ofAnnual
IEEE India Conference on (INDICON), Bangalore, India, 1–6.
Mohandes, M. A. 2015. Mobile technology for socio-religious events: A case study of NFC
technology. IEEE Technology and Society Magazine 34 (1):73–79. doi:10.1109/
MTS.2015.2396122.
Nguyen, H. K., and M. T. Chew. 2017. RFID-based attendance management system.
Proceedings of 2017 2nd Workshop on Recent Trends in Telecommunications Research
(RTTR), Palmerston North, New Zealand, 1–6.
Noguchi, S., M. Niibori, E. Zhou, and M. Kamada. 2018. Student attendance management
system with bluetooth low energy beacon and android devices. Proceedings of 18th
International Conference on Network-Based Information Systems, Taipei, Taiwan, 710–13.
Noor, S. A. M., N. Zaini, M. F. A. Latip, and N. Hamzah. 2016. Android-based attendance
management system. Proceedings of IEEE Conference on Systems, Process and Control
(ICSPC 2016), Melaka, Malaysia, 118–22.
Pedro, D. F., S. Tomic, L. Bernardo, M. Beko, R. Oliveira, R. Dinis, P. Pinto, and P. Amaral.
2019. Algorithms for estimating the location of remote nodes using smartphones. IEEE
Access 7:33713–27. doi:10.1109/ACCESS.2019.2904241.
Rana, A., and A. Ciardulli. 2017. Identity verification through face recognition: Android
smartphones and NFC. Proceedings of World Congress on Internet Security (WorldCIS-
2017), University of Cambridge, UK, 162–63.
Rekha, E., and P. Ramaprasad. 2017. An efficient automated attendance management system
based on Eigen face recognition. Proceedings of 2017 7th International Conference on
Cloud Computing, Amity University, India, Data Science & Engineering - Confluence,
605–08.
Shi, Y.-D., S.-H. Sim, and J.-C. Lee. 2017. Model-based integration of test and evaluation
process and system process for development of safety-critical weapon systems. Systems
Engineering 20 (3):120–38.
Sousa, J., and J. Barata. 2020. Tracking the wings of Covid-19 by modeling adaptability with
open mobility data. Applied Artificial Intelligence 35 (1):41–62. doi:10.1080/
08839514.2020.1840196.
Strnadelova, B., J. Halamova, and M. Kanovsky. 2019. Eye-tracking of facial emotions in
relation to self-criticism and self-reassurance. Applied Artificial Intelligence
33 (10):839–62. doi:10.1080/08839514.2019.1646004.
 APPLIED ARTIFICIAL INTELLIGENCE e2083796-1137

Sunehra, D., P. L. Priya, and P. L. Bano 2016. Children location monitoring on Google maps
using GPS and GSM technologies. Proceedings of IEEE 6th International Conference on
Advanced Computing (IACC), Bhimavaram, Andhra Pradesh, India, 711–15.
Takahashi, K., M. Niibori, and M. Kamada. 2021. A web-based attendance management system
with NFC seat identifiers. In Advances in Networked-Based Information Systems NBiS 2020
advances in intelligent systems and computing, L. Barolli, K. Li, T. Enokido, and M. Takizawa.
ed., vol. 1264, Cham: Springer, 576–581. doi:10.1007/978-3-030-57811-4_58.
Varadharajan, E., R. Dharani, S. Jeevitha, B. Kavinmathi, and S. Hemalatha. 2016. Automatic
attendance management system using face detection. Proceedings of Online International
Conference on Green Engineering and Technologies (IC-GET), Karpagam College of
Engineering, India, 1–3.
Walkhighlands. 2021 October. https://www.walkhighlands.co.uk/safety/gps-smartphones.
shtml.