Bojan Vasovic

Dr. Bojan Vasovic

Blace, Toplica Academy of Applied Studies, Prokuplje, Serbia
orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 February 2023 in Artificial Intelligence Review
Verified
Journal Article
Closed access
PDF file
SSSrećko Stamenković
Srećko Stamenković
NJNenad Jovanović
Nenad Jovanović
BVBojan Vasovic
Bojan Vasovic
MCMiloš Cvjetković
Miloš Cvjetković
ZJZoran Jovanović
Zoran Jovanović

1 January 2025
Verified
Conference Paper
Closed access
MIMilos Ilić
Milos Ilić
AIAleksandra Stojnev Ilic
Aleksandra Stojnev Ilic
MGMilan Grujev
Milan Grujev
SSSrećko Stamenković
Srećko Stamenković
BVBojan Vasovic
Bojan Vasovic
The integration of artificial intelligence in agriculture has significantly advanced data-driven decision-making across the sector. However, the "black box" nature of many AI models poses challenges in transparency, trust, and adoption, particularly among stakeholders with limited technical expertise. Explainable AI (XAI) addresses this limitation by making model predictions interpretable and understandable. This paper explores the application of XAI across seven critical agricultural domains: crop yield prediction, pest and disease detection, precision agriculture, soil health analysis, weather risk management, plant breeding and genomics, and market forecasting. For each application, we examine how XAI techniques—such as feature attribution, saliency maps, and rule-based explanations—can enhance interpretability, improve user confidence, and support actionable insights. Through case studies and model analyses, we demonstrate that incorporating explainability not only improves transparency but also leads to better-informed decisions, increased efficiency, and more sustainable farming practices. The findings highlight the transformative potential of XAI as a bridge between complex algorithms and practical agricultural use, promoting responsible and inclusive AI deployment in the agri-food sector.

30 June 2020 in Bizinfo Blace
Verified
Journal Article
Open access
PDF file
NJNenad Jovanovic
Nenad Jovanovic
BVBojan Vasovic
Bojan Vasovic
ZJZoran Jovanović
Zoran Jovanović
MCMiloš Cvjetković
Miloš Cvjetković
Artificial neural networks are a powerful tool that engineers can use in variety of purposes. The most common tasks are classification and regression, as well as control, modeling and prediction. For more than three decades, the field of artificial neural networks has been the center of intensive research. A large number of software tools have been developed to train these types of networks, but there is still interest in implementing neural networks in different programming languages. This paper aims to present the implementation of an arbitrary neural network in the Java programming language.

22 June 2026 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Dejan M. VasovićDejan M. Vasović
Dejan M. Vasović
GJGoran Janackovic
Goran Janackovic
ZVZarko Goran Vranjanac
Zarko Goran Vranjanac
SSSrećko Stamenković
Srećko Stamenković
BVBojan Vasovic
Bojan Vasovic
Educational information systems have evolved into highly interconnected digital landscapes that support learning management platforms, student information systems, institutional repositories, and online assessment environments. As these systems increasingly operate across cloud infrastructures and mobile devices, ensuring the confidentiality, integrity, and availability (CIA Triad) of educational data is critical for safeguarding institutional operations and maintaining trust in digital education services. This paper investigates how next-generation security protocols, such as adaptive multi-factor authentication and advanced access control and data protection mechanisms, can reinforce ISO/IEC 27001:2022 requirements within contemporary educational information systems. The analysis maps emerging protocol capabilities to relevant new ISO/IEC 27001:2022 control domains, illustrating how they mitigate threats associated with unauthorized access, data manipulation, and service disruption. The proposed framework is further supported by an implementation-oriented mapping and an illustrative operational architecture that demonstrates the feasibility of translating prioritized security determinants into practical mechanisms. The FAHP analysis identifies access control mechanisms, backup and recovery, and data validation as the three highest-weighted determinants, with aggregate weights of 0.061, 0.059, and 0.057, respectively. These determinants are translated into a determinant-driven Security Operationalization Matrix that connects ISO/IEC 27001:2022 control domains, CIA dimensions, and technology recommendations, and is complemented by implementation feasibility considerations tailored to the budgetary, infrastructural, and resource constraints characteristic of educational institutions. Based on the prioritization results and conceptual operationalization, the proposed integrative approach provides a structured and progressively adoptable foundation for CIA-oriented security governance in digital educational environments.

1 January 2025
Verified
Conference Paper
Closed access
SSSrećko Stamenković
Srećko Stamenković
MIMiloš Ilić
Miloš Ilić
BVBojan Vasovic
Bojan Vasovic
ZJZoran Jovanović
Zoran Jovanović
Ivan GarvanovIvan Garvanov
Ivan Garvanov
In modern education, the development of educational software requires careful alignment of different goals in order to ensure an effective, pedagogically relevant and technically stable learning experience. This paper aims to explore and systematize taxonomic approaches applied in the design of educational software systems. Based on the analysis of relevant literature in the fields of pedagogy, software engineering and digital technologies in learning, a three-dimensional taxonomy of goals is formulated, which includes: (1) technical goals, (2) pedagogical goals, and (3) usability goals. This structure enables a holistic approach to the design of educational tools and provides a basis for evaluating and improving existing systems. The literature review indicates the fragmentation of existing taxonomies and the need to unify pedagogical and engineering principles into a single framework. The proposed taxonomy represents a step towards standardization and improvement of educational software design in accordance with modern educational needs.

1 January 2025
Verified
Conference Paper
Open access
PDF file
BVBojan Vasovic
Bojan Vasovic
ZJZoran Jovanović
Zoran Jovanović
VMVioleta Milićević
Violeta Milićević
This study aims to identify and systematize contemporary factors associated with IT project failure and examine their relationship with factors established in earlier literature. A systematic literature review covering publications from January 1, 2021, to September 30, 2025, was conducted in accordance with PRISMA 2020 and PRISMA-S guidelines. Fifteen studies were included in the qualitative synthesis, while 12 studies were used for factor and frequency analysis. Data extraction yielded 86 coded findings, which were standardized into 76 distinct factors and grouped into seven dimensions. Project management and governance was the most prevalent dimension, appearing in 75.0% of the studies, while inadequate planning was the most frequently identified individual factor. Comparison with literature published up to 2020 showed that 65.8% of the factors were traditional or persistent, 27.6% were evolving, and 6.6% were contemporary. Contemporary factors were primarily associated with AI/ML projects and risks related to data, privacy, and model maintenance. The findings indicate that the digital era has not eliminated traditional causes of IT project failure, but has increased their complexity and interdependence.