Nikola Gligorijević

International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE)

A Hybrid SWARA-NWA Framework for Evaluating AI-Based Image Recognition Algorithms in Educational Technology Applications

Gligorijević N., Viduka D., Đukić-Popović S., Nikolić V., Popović S.

https://ijcrsee.com/index.php/ijcrsee/article/view/3377

Artificial Intelligence (AI) and computer vision technologies are increasingly integrated into educational environments through intelligent tutoring systems, gesture-based learning, facial expression analysis, and automated evaluation tools. However, selecting the most appropriate image recognition algorithms for educational applications remains a challenge due to varying requirements regarding speed, accuracy, hardware compatibility, and usability in dynamic classroom conditions. This paper proposes a hybrid multi-criteria decision-making (MCDM) model based on the Step-wise Weight Assessment Ratio Analysis (SWARA) and Net Worth Analysis (NWA) methods to evaluate and rank nine widely used AI-based visual recognition algorithms. The evaluation is conducted using five education-relevant criteria: processing speed, recognition accuracy, robustness to classroom noise, compatibility with low-end devices, and energy efficiency. Expert assessments from the field of educational technology were used to derive weight coefficients and evaluate algorithm performance. The results show that Fast R-CNN achieved the highest overall score (1.141), followed by U-Net (1.077) and DeepLab (1.062), indicating their suitability for real-time and resource-constrained EdTech environments. Algorithms such as MobileNet (1.057) and YOLO (1.037) also demonstrated balanced performance, making them viable for mobile or moderately demanding educational scenarios. The proposed model offers a structured and transparent decision-support framework that can assist researchers and practitioners in selecting optimal AI algorithms for diverse educational applications.

SWARA, NWA, artificial intelligence, image recognition, educational technology

M22 – Rad u međunarodnom časopisu

Advances in Electrical and Electronic Engineering (AEEE)

Performance Evaluation of Hybrid RF/FSO Systems Using Nakagami-M and Gamma-Rician Models for One-Hop and Multi-Hop Scenarios

Đorđe Šarčević, Nenad Stanojević, Stefan Panić, Petar Spalević, Nikola Gligorijević, Čedomir Vasić

—

In this study, a comprehensive model for evaluating the performance of hybrid RF/FSO (Radio Frequency/Free Space Optics) systems is introduced. The recently presented Gamma-Rician model was employed for the statistical characterization of FSO turbulence, while the Nakagami-m fading model was utilized for RF propagation modeling. We provide closed-form expressions for the Cumulative Distribution Function (CDF) of observed Signal-to-Noise Ratio (SNR) at reception, along with closed-form expressions for the Average Bit Error Rate (ABER) using the Coherent Binary Phase Shift Keying (CBPSK) modulation scheme. Two transmission methods in relay hybrid RF/FSO systems have been analyzed: One-Hop and Multi-Hop transmissions. By capitalizing on these expressions, we have evaluated the ABER performance of these systems as a function of various parameters, highlighting the robustness of hybrid configurations under differing operational conditions.

Gamma-Rician model, Hybrid RF/FSO systems, Multi-Hop transmission, Nakagami-m fading

M23 – Rad u međunarodnom časopisu

Journal of Business Economics and Management

Multi-criteria Decision Analysis of E-commerce Software Selection Using AHP-NWA Hybrid Model

Rakić R., Gligorijević M., Viduka D., Gligorijević N., Strugarević D.

https://journals.vilniustech.lt/index.php/JBEM/article/view/25406

In the digital economy, selecting the right e-commerce platform is a strategic decision with significant implications for efficiency and competitiveness. This paper applies a hybrid decision-making framework that integrates the Analytic Hierarchy Process (AHP) and Net Worth Analysis (NWA) to evaluate five popular e-commerce platforms: Magento, WooCommerce, Shopify, PrestaShop, and OpenCart. AHP was used to derive weights for evaluation criteria, while NWA incorporated expert assessments of alternatives. Results indicate that security (32%) and functionality (25%) are the most critical factors, followed by maintenance costs (14%) and scalability (11%). The ranking shows Magento as the leading platform (0.575), excelling in security and functionality, while WooCommerce (0.567) is highly flexible and Shopify (0.563) stable though less customizable. PrestaShop (0.505) and OpenCart (0.496) scored lower, making them suitable for smaller businesses. The contribution of this study lies in the integration of AHP-derived weights into the NWA framework under a dual expert panel structure, ensuring methodological independence and reducing bias. This hybrid approach offers both practical implications for digital business strategy and theoretical insights into combining hierarchical and network-based MCDM methods, thereby addressing a research gap in e-commerce software evaluation.

e-commerce, multi-criteria decision-making, AHP, NWA, software platforms, Magento, WooCommerce, Shopify

M21 – Rad u vodećem međunarodnom časopisu

Scientific Journal Management in Sports

INTEGRACIJA VEŠTAČKE INTELIGENCIJE U ISTRAŽIVAČKI PROCES SPORTSKOG MENADŽMENTA: POTENCIJALI I IZAZOVI

Milojević S., Gligorijević N., Savičević D., Milašinović M.

https://doi.org/10.46793/ManagSport16-2.156

Artificial intelligence (AI) is becoming an increasingly important tool in scientific research, including in the field of sports management. This paper provides an overview of how AI can be applied in all phases of scientific research – from defining problems and goals, through literature review and hypothesis formulation, to research design and sampling, data collection, analysis, hypothesis testing, interpretation of results, and dissemination of findings. Special attention is paid to examples of the application of AI in various areas of sports management, such as strategic management and organizational culture, sports marketing, finance and economics of sports, human resource management and leadership, sports law and policy, ethics and integrity in sports, management of sports events and facilities, technology and innovation in sports, sports analytics, sustainable development in sports, and international sports management. Through a descriptive analysis of relevant literature and case studies, the paper highlights how AI contributes to the efficiency of the research process (e.g., faster data processing, pattern detection, and insight generation) in sports management, while also discussing challenges such as validity, algorithm bias, and ethical dilemmas. The results suggest that integrating AI can enhance the quality and impact of scientific research in sports management, provided that a critical approach is maintained and scientific integrity is maintained. The paper concludes that the ability of researchers to utilise AI tools, while understanding their limitations effectively, will be crucial for the future development of sports management science.

artificial intelligence, scientific research, sports management, research methodology, digital transformation of scientific research

M24

International Scientific and Professional Conference “ALFATECH” Smart Cities and Modern Technologies

Integration of Drones for Intelligent Crowd Counting in the Safe City Concept

Gligorijević N., Strugarević D., Čabrić V., Račić M.

https://unicloud.kg.ac.rs/index.php/s/sGGRfeWQ3k4tDay?dir=/&editing=false&openfile=true

Rapid urbanization and increasing population density in urban areas pose a significant challenge to maintaining security and effective crowd management during mass gatherings, such as public events, protests and emergencies. The "safe city" concept relies on modern technologies, including drones and artificial intelligence, to improve security, optimize resource allocation and reduce the risks associated with mass gatherings. This paper explores the use of drones, equipped with advanced cameras and object detection algorithms like YOLO and Fast R-CNN, to count people in crowds and analyze their movements in real time. By combining multi-criteria analysis, the algorithms were evaluated according to key criteria, including accuracy, processing speed, robustness to noise, segmentation efficiency and energy efficiency. The results show that the YOLO algorithm is superior in applications that require fast real-time processing, while Fast R-CNN provides higher accuracy in complex scenarios. Integrating drones with these algorithms enables accurate crowd counting and tracking, which contributes to better security and management in modern urban environments.

safe city, drones, artificial intelligence, YOLO, Fast R-CNN, crowd counting, multi-criteria analysis

M33

15th European Symposium on Computational Intelligence and Mathematics, May 12th–15th, 2024, Krakow, Poland

Using Human-Computer Interaction Data for Continuous Authentication in High-stake Electronic Assessments

Strugarević D., Gligorijević N., Šimić G., Jevremović A.

https://escim2026.uca.es/wp-content/uploads/Book_of_abstracts_ESCIM_2024.pdf#page=57

As electronic assessments become more prevalent in high-risk and online environments, ensuring the reliability of user authentication becomes increasingly important. In this research we examine the potential for using human-computer interaction (HCI) data for improving the reliability of authentication process in electronic assessments. By analyzing user interaction patterns with assessment interfaces, we try to improve the robustness of the authentication system. Our approach uses machine learning algorithms to discern subtle behavioral signals, primarily response times, to establish a user profile. This profile is then compared against the profile claimed within the primary authentication. The main characteristic of the interaction we used in this initial research is response time. The results indicate that the data collected from participants during a typical test (49 participants and 35 questions with provided answers, in the Health Management course) are not sufficient to enhance authenticity verification as proposed. Therefore, it is necessary to develop a more comprehensive profile of participants for such an approach to make sense.

Human-Computer Interaction, Authenticity, High-stake Electronic Assessments

M33

In the Book of the Proceedings of IRASA International Scientific Conference, Science, Education, Technology and Innovation, SETI VII

Methodological Approach to Educating Students in the Field of Marketing in the Age of Artificial Intelligence

Brkljač, M., Gligorijević, M., Gligorijević, N.

https://doi.org/10.62982/seti07.mibr.28

The dynamic nature of marketing development and the contemporary influence of information communication technologies on most marketing strategies and tactics have led to the need to review traditional learning methods in the field of marketing. This paper aims to present an analysis of existing literature that deals with methodological approaches to education in the field of marketing, as well as to present critical observations and suggestions by emphasizing the relevance of applied methods in the context of the use of artificial intelligence tools. Emphasis is placed on those learning models that relate to the development of key skills of future marketing experts, such as critical thinking, analytical skills and creativity, as opposed to models that aim at pure reproduction of mastered material. The learning models analysed in this paper include PBL - problem-based learning, project-based learning, the use of simulations, as well as the integration of AI-based tools (adaptive learning, interactive chatbots, and predictive analytics). The necessity for students is obvious to move from reactive to proactive learning methods. In the paper, an attempt was made to point out the shortcomings of the existing literature and to give suggestions for future research in the field of educational policies.

Methodology, marketing education, learning models, AI based tools, proactive learning, critical thinking and innovation

M33