GPS-based Classification Algorithm for Employee Attendance System using Telegram API

Citation

Taqi, M. K. (2026). GPS-based Classification Algorithm for Employee Attendance System using Telegram API. International Journal of Research, 13(1), 406–415. https://doi.org/10.26643/ijr/2026/14

Mustafa Kadhim Taqi
Technical College of Management – Kufa, Al-Furat Al-Awsat Technical University, Kufa, 54003, Iraq

Email: ktmustafa@atu.edu.iq

Abstract

The attendance system for employees, which is mostly used across the globe, is based on a fingerprint device. The drawbacks of this system are the presence of tool dependency, lower availability of fingerprint scanners, and the equipment being far away from the work premises. Due to the mentioned shortcomings, we propose an application system for presence built on the Telegram Bot using GPS. It will aid the employee in showing up in their work area. By installing the proposed system, numerous benefits will result. It will ease the overall presence system, and the processing of data on presence will be much more automated and easier. Due to the Telegram Bot method, the system can easily navigate the employee data, highlight daily attendance output, and efficiently store the presence results. It has a prediction accuracy of 87.5%, an acquired system sensitivity of 80%, and a shown specificity of about 91%.

Keywords: Attendance system, Telegram BOT, Classification, GPS.

1.  Introduction

The dire duty of the employee is to be present in their workspace [1]. Employee discipline can be measured from the presence system by evaluating the presence data. The presence data approach is from marking attendance. Numerous ways are utilized in the procedure of obtaining presence data, i.e., fingerprint, signature, and scanned barcode [2]. Most organizations are using handprints or fingerprints as their standard presence method [3].

Fingerprints as a presence method is one of the most renowned ways of obtaining presence data [4]. This method reduces the fraud ratio as each individual has a unique fingerprint. The deployment of fingerprint sensors is mostly scarce and limited. The employee has to move to the specific space to mark their presence, where the equipment is installed [5]. Rather, the employee doesn’t need to be close to the area of the fingerprint attendance marking equipment. Though they are physically present in the organization when they are in their working area. It can be summed up that employees show presence at their workplace [6]. To aid the employees who work not so close to the presence marking equipment, an innovative presence system is required [7]. Through that unique presence system, presence data can be collected and obtained from any space of the actual work environment. The employee can record their presence from anywhere on the work premises. Such an employee attendance model can be developed through GPS-based using a Telegram Bot [8].

• Methods and Materials

An attendance methodology is a way that is employed for storing, scrutinizing, and obtaining a screenshot of the attendance profile of each organizational member. The purpose of the attendance system is to store the presence of each person along with their time of arrival and departure. For conducting this research study, different research methodologies were used. These include:

• Database design
• Telegram BOT design
• API design
• System analysis and testing

            Generally, the application is developed to mark the presence of employees. In this application, each member verifies their presence via a Telegram BOT, which is present in a designated area in the organization. The data is then sent to the installed server. API receives the data, and then it is stored in the database. A Telegram BOT has been developed that can be accessed by the department admin. Through the application, they can preview data on presence that has been stored in the database. The process is illustrated in Figure 1.

Figure 1. Telegram BOT architecture for employee attendance systems using GPS

• Entity-Relationship Diagrams (ERD)

The proposed database has been designed by employing ERD. ERD can classify the required needs for the database among constructing systems [9]. A detailed illustration can be viewed in the diagram below. The diagram depicts six tables termed userstb, rolestb, users_rolestb, attendances, locations, and shift tables. The primary-foreign relationship between the userstb and users_rolestb tables has been established based on the user identity number (user_id). On the other hand, the relationship between the userstb, attendance, and shift tables has been established based on the chat_id given by the Telegram BOT. locations and attendance tables were related to the location ID.

Figure 2. Database design using entity-relationship diagrams (ERD).

• The Design of the Telegram BOT

It is composed of a thorough design pertinent to the involved users, flow, and roles of the entire system. It also includes the user interface design. Only the admin can access the Telegram BOT’s menu for controlling the attendance system. In that menu, the admin can view allowed attendance locations, delete locations, edit locations’ data, and even add new locations in the use case diagram. Figure 3 illustrates how the admin adds a new attendance registration location.

(a) System settings menu	(b) Add a new attendance location
(c) Read GPS location command	(d) Adding confimation message

Figure 3. Admin menu for the GPS-based employee attendance system

The employee menu comprises two items for attendance registration. The first menu item is for presence registration at the beginning of the shift. The second item is dedicated to dismissing registration. Figure 4 illustrates the employee menu items and the process of presence registration.

(a) Employee menu	(b) Presence and dismiss menu
(c) Command for location registration	(d) Share employee location (GPS)
(e) Accepting presence	(f) Rejecting presence

Figure 4. Employee menu for GPS-based employee attendance system

The attendance system of employees has its basis in the GPS of the Telegram Bot, upon which the API (Application Programming Interface) is developed. API acts as an intermediary among systems of data communication that are present on the server. It has an application on Telegram Bot. The involvement of API speeds up the procedure of designing applications on the Telegram Bot as the API gives the needed features. Due to this feature of API, the developers do not have to add parallel features. Figure 5 shows the test location point.

Figure 5. Test points

• Obtaining Data. The API design initiates when it obtains data in the format of the longitude and latitude of the device. The sequence is checked in as location and data completeness.
  • Developing API register. The development of REST API initiates after the API starts obtaining data in the pattern of birthplace and parent number. Then, checking in sequence, termed employment status data completeness, and employee data for more clarity.
  • Use Case Diagrams Design. Telegram Bot applications can be used by authorized users and employees. These features are available, i.e., attendance and register.
  • Designing the activity diagram. The Telegram BOT is developed with 2 core features, i.e., presence and registration. In the registration menu, the app instantly requests data from the IMEI device on the Telegram BOT. This data is sent to the server, which is tallied with the database.

The application attendance menu prompts for GPS data [7], and IMEI device data. If the GPS location is valid and a success then the collected data will be transferred to the server database. As result the server will transfer a failed or successful response to mark the presence and then it will preview it on the Telegram BOT application.

•  Outcomes

While conducting the test prior, the user can state the place/location for system testing. The testing location points and figures are outlined below. Different testing points at various locations present outside and inside are employed. Figure 8 depicts the testing and the outcomes are recorded in Table 1. There are multiple provisions in the test outcomes termed as:

• True positives: presence is classified as inside an area
• True negatives: presence is classified as inside in the outside area.
• False positives: area presence is termed outside.
• False negatives: presence is termed as outside.

Table 1. Results of presence at several attendance registration points

Point	Latitude	Longitude	Presence (Inside/Outside)	Provisions
1	32.02019	44.24388	Outside	TN
2	32.033908	44.410864	Outside	TN
3	32.033947	44.410949	Outside	TN
4	32.033567	44.411552	Inside	FP
5	32.033618	44.411373	Outside	FN
6	32.033997	44.411006	Outside	TN
7	32.033901	44.411042	Outside	TN
8	32.033536	44.411483	Inside	TP
9	32.033576	44.411512	Inside	TP
10	32.033952	44.411119	Outside	TN
11	32.033797	44.411142	Outside	TN
12	32.033772	44.411258	Outside	TN
13	32.033612	44.411482	Inside	TP
14	32.033605	44.411427	Inside	TP
15	32.033732	44.411262	Outside	TN
16	32.033746	44.411178	Outside	TN

From the table above it is found that the values of TP = 4, TN = 10, FP = 1, and FN = 1, the values of sensitivity, specificity, and accuracy of the system are as follows:

• Conclusions

The outlined method developed for the employee presence system is composed of 11 functions that are operating smoothly. Registration REST API and REST API attendance developed for employee attendance systems can interact with systems on the Telegram BOT. The proposed system has a sensitivity of 80%, a specificity of 91%, and an accuracy rate of 87.5%, demonstrating that the system is successfully running.

However, it is worth mentioning that the system may not grasp the precise location inside the concrete buildings, which may explain the fourth test point where the system incorrectly predicts it. On the other hand, the presence points close to the desired registration points by the admin may also be incorrectly predicted. This case has been shown with the fifth test point.

Based on the obtained results, the author recommends using the proposed system in registering the presence of employees.

References

[1]        Setiowati, R., et al., Development of Employees Attendance Features of Human Resource Information System in A National Logistics Company. 2023: p. 136-140.

[2]        Nasution, T.H., et al., Design of Portable Fingerprint System Prototype for Student Presence Integrated with Academic Information System at the Universitas Sumatera Utara. 2019. 1(1): p. 47-54.

[3]        Ekowati, V.M., et al., An Empirical Approach to Evaluate Employee Performance Using Finger Print Attendance. 2024. 25(199): p. 57-64.

[4]        Rahkoyo, E., et al. IoT-Based Fingerprint Attendance System: Enhancing Efficiency and Security in Educational and Organizational Settings. in 2024 International Conference on Advances in Modern Age Technologies for Health and Engineering Science (AMATHE). 2024. IEEE.

[5]        Abbas, Z., et al., A Fingerprint based Students attendance System with SMS alert to Parents. 2023. 6(3).

[6]        El-Mawla, A., et al., Smart Attendance System Using QR-Code, Finger Print and Face Recognition. 2022. 2(1): p. 1-16.

[7]        Singh, R., et al. Fingerprint-based Portable Attendance Monitoring System using Raspberry Pi Pico. in 2024 11th International Conference on Computing for Sustainable Global Development (INDIACom). 2024. IEEE.

[8]        Dali, S.W., H. Hadiwiyatno, and D.W.J.J.o. T.N. Illahi, Design and development of a web-based compensation information and registration system using biometric fingerprint approval delivery using telegram bot digital: registration and submission of compensation using the fingerprint and telegram websites. 2023. 13(4): p. 385-394.

[9]        Pulungan, S.M., et al., Analisis Teknik Entity-Relationship Diagram Dalam Perancangan Database. 2023. 1(2): p. 143-147.