1. A secure way of communication for smart grid networksPooja Tyagi, Saru Kumari, Mohammed J. F. Alenazi, Marko Hölbl, 2025, original scientific article Abstract: In 2021, Khan et al. suggested a scheme based on smart grid networks. They deployed the random oracle model to justify the security of their scheme formally. They also verified the security of the scheme using the AVISPA software tool. We studied this scheme and found some security issues. The scheme suffers from a confidentiality breach attack. In Khan et al.’s scheme, an adversary can track both the user and the server, and it does not provide user and server anonymity. An adversary can also impersonate a server. To overcome all these security issues, we propose a protocol for smart grid networks. In our proposed scheme, we maintain all the qualities of Khan et al.’s scheme and try to remove all its weaknesses.
Keywords: key agreement, smart grid, smart grid networks, user authentication
Published in DKUM: 01.12.2025; Views: 0; Downloads: 0
 Full text (1,71 MB) |
2. A machine vision approach to assessing steel properties through spark imagingGoran Munđar, Miha Kovačič, Uroš Župerl, 2025, original scientific article Abstract: Accurate and efficient evaluation of steel properties is crucial for modern manufacturing. This study presents a novel approach that combines spark imaging and deep learning to predict carbon content in steel. By capturing and analyzing sparks generated during grinding, the method offers a fast and cost-effective alternative to conventional testing. Using convolutional neural networks (CNNs), the proposed models demonstrate high reliability and adaptability across different steel types. Among the tested architectures, MobileNet-v2 achieved the best performance, balancing accuracy and computational efficiency. The findings highlight the potential of machine vision and artificial intelligence in non-destructive steel analysis, providing rapid and precise insights for industrial applications.
Keywords: carbon content prediction, convolutional neural networks, deep learning, machine vision, spark imaging, steel analysis
Published in DKUM: 03.11.2025; Views: 0; Downloads: 5
Full text (1,84 MB)
This document has many files! More... |
3. The impact of artificial intelligence in business communication on social networks: a case studyBlaž Kovač, 2025, undergraduate thesis Abstract: Artificial intelligence has revolutionised the way we communicate in recent years, especially on social networks, where communication is fast and large-scaled. Companies are also using artificial intelligence to communicate with customers, including automating messages intended for customer acquisition, which raises a key question: how effective are messages created with artificial intelligence compared to messages written without artificial intelligence? This thesis analyses the impact of artificial intelligence on business communication using the example of company X. The research compares the effectiveness of messages created with artificial intelligence with messages written without artificial intelligence in the form of LinkedIn outreach, focusing on two key performance indicators: the response rate to the messages and the number of video calls made with potential customers. In addition, the time efficiency of preparing content for messages with and without artificial intelligence was also analysed.
The results show that messages created with artificial intelligence were not only faster to prepare but also achieved higher response rates and a higher number of scheduled video calls with customers compared to messages written without artificial intelligence. This thesis contributes to a better understanding of the role of artificial intelligence in modern business communication and offers insights into its value in connecting with potential customers via social networks, using a practical example.
Keywords: Artificial intelligence, business communication, social networks, comparison.
Published in DKUM: 03.10.2025; Views: 0; Downloads: 18
Full text (781,87 KB) |
4. Micro-location temperature prediction leveraging deep learning approachesAmadej Krepek, Iztok Fister, Iztok Fister, 2025, original scientific article Abstract: Nowadays, technological progress has promoted the integration of artificial intelligence into modern human lives rapidly. On the other hand, extreme weather events in recent years have started to influence human well-being. As a result, these events have been addressed by artificial intelligence methods more and more frequently. In line with this, the paper focuses on searching for predicting the air temperature in a particular Slovenian micro-location by using a weather prediction model Maximus based on a longshort term memory neural network learned by the long-term, lower-resolution dataset CERRA. During this huge experimental study, the Maximus prediction model was tested with the ICON-D2 general-purpose weather prediction model and validated with real data from the mobile weather station positioned at a specific micro-location. The weather station employs Internet of Things sensors for measuring temperature, humidity, wind speed and direction, and rain, while it is powered by solar cells. The results of comparing the Maximus proposed prediction model for predicting the air temperature in micro-locations with the general-purpose weather prediction model ICON-D2 has encouraged the authors to continue searching for an air temperature prediction model at the micro-location in the future.
Keywords: long short-term memory neural networks, air temperature, micro-location, prediction, weather, Internet of Things
Published in DKUM: 25.09.2025; Views: 0; Downloads: 10
Full text (8,81 MB) |
5. Support structures and intergenerational support during and after the COVID-19 pandemicDunja Potočnik, Andrej Naterer, 2025, independent scientific component part or a chapter in a monograph Abstract: This chapter examines the role of formal and informal support structures in shaping the well-being and resilience of youth in Croatia and Slovenia. In both countries, families remain the most important support system, particularly mothers, who are consistently identified as central figures in providing emotional and practical assistance. While peers also play a crucial role, the pandemic disrupted these relationships and reduced opportunities for in-person interaction. Institutional support, such as educational and employment services, remains important but often perceived as inaccessible or poorly adapted to the actual needs of youth. At the same time, a low level of trust in political institutions and the welfare system was observed, particularly in Croatia, which reinforces reliance on familial networks. Digital platforms increasingly serve as alternatives for connection and advice, although they cannot replace interpersonal support. Policy implications stress the need to expand accessible, youth-centred services, including mental health care, career guidance, and community-based initiatives. Strengthening institutional trust and investing in participatory frameworks would help diversify support beyond families and foster more resilient pathways for young people's social integration and life transitions.
Keywords: youth support, family networks, institutional trust, mental health, Croatia and Slovenia
Published in DKUM: 09.09.2025; Views: 0; Downloads: 6
Full text (828,67 KB)
This document has many files! More... |
6. Ensemble-based knowledge distillation for identification of childhood pneumoniaGrega Vrbančič, Vili Podgorelec, 2025, original scientific article Abstract: Childhood pneumonia remains a key cause of global morbidity and mortality, highlighting the need for accurate and efficient diagnostic tools. Ensemble methods have proven to be among the most successful approaches for identifying childhood pneumonia from chest X-ray images. However, deploying large, complex convolutional neural network models in resource-constrained environments presents challenges due to their high computational demands. Therefore, we propose a novel ensemble-based knowledge distillation method for identifying childhood pneumonia from X-ray images, which utilizes an ensemble of classification models to distill the knowledge to a more efficient student model. Experiments conducted on a chest X-ray dataset show that the distilled student model achieves comparable (statistically not significantly different) predictive performance to that of the Stochastic Gradient with Warm Restarts ensemble method (F1-score on average 0.95 vs. 0.96, respectively), while significantly reducing inference time and decreasing FLOPs by a factor of 6.5. Based on the obtained results, the proposed method highlights the potential of knowledge distillation to enhance the efficiency of complex methods, making them more suitable for utilization in environments with limited computational resources.
Keywords: knowledge distillation, convolutional neural networks, childhood pneumonia
Published in DKUM: 20.08.2025; Views: 0; Downloads: 7
Full text (915,47 KB) |
7. Contextualized spatio-temporal graph-based method for forecasting sparse geospatial sensor networksNiko Uremović, Domen Mongus, Aleksander Pur, Niko Lukač, 2025, original scientific article Abstract: Spatio-temporal forecasting is a rapidly evolving field, accelerated by the increasing accessibility of sensoring infrastructure and computational hardware, capable of processing the large amount of sampled data. Applications of spatio-temporal forecasts range from traffic, weather, air pollution forecasting and others. Emerging technologies employ deep learning architectures, such as graph, convolutional, recurrent and transformer neural networks. While the state-of-the-art methods provide accurate time series predictions, they are typically limited to providing forecasts only for the direct locations of sampling, whereas coverage of the entire area is often desired by the applications. In this work, we propose a method that addresses this challenge and improves on the shortcomings of related works, which have already tackled the task. The proposed graph convolutional recurrent neural network based method provides forecasts for arbitrary geolocations without available measurement data, formulating predictions based on contextual information of target geolocations and the time series data of nearby measurement geolocations. We evaluate the method on three real-world datasets from meteorological, traffic and air pollution domains, and gauge its performance against the state-of-the-art spatio-temporal forecasting methods. The proposed method achieves 12.26 %, 66.97 % and 42.89 % improvements in the mean absolute percentage errors on the three aforementioned datasets, compared to the best performing state-of-the-art method GConvGRU.
Keywords: spatio-temporal forecasting, graph recurrent neural networks, sparse geospatial sensor networks
Published in DKUM: 25.07.2025; Views: 0; Downloads: 2
Full text (5,19 MB) |
8. Optimal ensemble-based framework for ground-fault protection in radial MV distribution networks with resonant grounding☆Boštjan Polajžer, Younes Mohammadi, Thomas Olofsson, Gorazd Štumberger, 2025, original scientific article Abstract: Ground fault relays (GFRs) in resonant-grounded medium voltage distribution networks shall not operate during phase-to-ground (Ph-G) fault inception, allowing the Petersen coil to suppress self-extinguishing faults, but the designated GFR must operate during permanent faults. In order to enhance the performance of GFRs, particularly during high-impedance faults, the scope of this paper is to propose a straightforward, machine-learning-based protection framework. The enhanced GFR is modeled as a classification task. Depending on the GFR’s position and the Ph-G fault location in the network, fault samples are labeled as “no operation,” “primary,” “backup,” or “backup of backup,” forming two-class, three-class, and four-class GFR setups, respectively. This assures selective operation across three protection zones and improves the reliability of all GFRs. The proposed protection scheme employs backward optimal feature selection to identify the most relevant discrete features obtained from measured zero-sequence current and voltage waveforms. An ensemble of k-nearest neighbor classifiers is utilized for accurate classification, simulating the GFR operating conditions, with measurement errors and sensitivity incorporated in the preprocessing. A 20 kV case study network validates the proposed framework, achieving F1-scores exceeding 96 %. The maximum operation delay of the protection scheme for an enhanced GFR is 225 ms, accommodating the required time window (200 ms), prediction time (5 ms), and change detection time (20 ms), thus assuring safe operation. Compared to other machine-learning-based methods used for Ph-G fault protection in resonant-grounded radial networks, this framework is high-performing, fast, and easy to implement, utilizing a simpler structure than neural networks.
Keywords: resonant grounded networks, ground-fault relay, high-impedance faults, ensemble-based learning, optimal feature selection
Published in DKUM: 25.07.2025; Views: 0; Downloads: 4
Full text (4,86 MB) |
9. Synchronization stability in simplicial complexes of near-identical systemsFatemeh Parastesh, Mahtab Mehrabbeik, Karthikeyan Rajagopal, Sajad Jafari, Matjaž Perc, Charo I. del Genio, Stefano Boccaletti, 2025, original scientific article Abstract: Assessing the stability of synchronization is a fundamental task when studying networks of dynamical systems. However, this becomes challenging when the coupled systems are not exactly identical, as is al ways the case in practical settings. Here we introduce an extension of the Master Stability Function to determine near-synchronization stability within simplicial complexes of nearly identical systems coupled by synchronization-noninvasive functions. We validate our method on a simplicial complex of Lorenz oscillators, f inding a good correspondence between the predicted regions of stability and those observed via direct simula tion. This confirms the correctness of our approach, making it a valuable tool for the evaluation of real-world systems, in which differences between the constitutive elements are unavoidable.
Keywords: chaos, collective dynamics, coupled oscillators, dynamics of networks, synchronization, chaotic systems, dynamical systems, networks
Published in DKUM: 10.07.2025; Views: 0; Downloads: 9
Full text (2,72 MB)
This document has many files! More... |
10. Meta analysis of business valuation solutions –are AI based methods better?Aljaž Herman, Damijan Mumel, Timotej Jagrič, 2024, review article Abstract: Purpose of the article–this article addresses the challenge of accurately assessing business value in today's dynamic environment, exploring the limitations of traditional valuation methods and the potential of modern, technology-driven approaches.Research methodology–the study uses qualitative research methods, including content analysis, deductive reasoning, and comparative analysis, to review various business valuation techniques.Findings –the research finds that traditional methods like Discounted Cash Flow and Relative Valuation are outdated, failing to capture all value factors. Modern approaches, such as simulation-based valuation, machine learning, and neural networks, combine traditional methods with advanced techniques. These methodologies utilize vast datasets and sophisticated algorithms, enhancing predictive accuracy and understanding of market dynamics. Neural networks excel in analysing complex patterns and adapting to market shifts. However, no single method can capture all nuances, necessitating diverse approaches and acknowledging the subjective nature of valuations
Keywords: business valuation, traditional and advanced valuation methods, machine learning, neural networks, artificial intelligence
Published in DKUM: 01.07.2025; Views: 0; Downloads: 9
Full text (577,12 KB)
This document has many files! More... |
