1. Data protection heterogeneity in the European UnionMarko Hölbl, Boštjan Kežmah, Marko Kompara, 2021, original scientific article Abstract: In light of digitalisation, we are witnessing an increased volume of collected data and
data generation and exchange acceleration. Therefore, the European Union (EU) has introduced the
General Data Protection Regulation (GDPR) as a new framework for data protection on the European
level. However, GDPR allows the member states to change some parts of the regulation, and the
member states can always build on top of the GDPR. An example is the collection of biometric data
with electronic signatures. This paper aims to compare the legislation on data protection topics in the
various EU member states. The findings show that the member states included in the study generally
do not have many additional/specific laws (only in 29.4% of the cases). However, almost all have
other/additional legislation to the GDPR on at least one topic. The most additional legislation is on
the topics of video surveillance, biometry, genetic data and health data. We also introduce a dynamic
map that allows for quick navigating between different information categories and comparisons of
the EU member states at a glance. Keywords: data protection, personal data, data privacy, GDPR legislation, heterogeneity, legislation on data protection, European Union Published in DKUM: 16.06.2025; Views: 0; Downloads: 0
Full text (462,91 KB) This document has many files! More... |
2. Synergy of blockchain technology and data mining techniques for anomaly detectionAida Kamišalić Latifić, Renata Kovačević, Iztok Fister, 2021, review article Abstract: Blockchain and Data Mining are not simply buzzwords, but rather concepts that are playing
an important role in the modern Information Technology (IT) revolution. Blockchain has recently
been popularized by the rise of cryptocurrencies, while data mining has already been present in IT
for many decades. Data stored in a blockchain can also be considered to be big data, whereas data
mining methods can be applied to extract knowledge hidden in the blockchain. In a nutshell, this
paper presents the interplay of these two research areas. In this paper, we surveyed approaches for
the data mining of blockchain data, yet show several real-world applications. Special attention was
paid to anomaly detection and fraud detection, which were identified as the most prolific applications
of applying data mining methods on blockchain data. The paper concludes with challenges for future
investigations of this research area. Keywords: anomaly detection, blockchain, distributed ledger, data mining, machine learning Published in DKUM: 16.06.2025; Views: 0; Downloads: 0
Full text (681,73 KB) This document has many files! More... |
3. Feasibility of a computerized clinical decision support system delivered via a socially assistive robot during grand rounds : a pilot studyValentino Šafran, Urška Smrke, Bojan Ilijevec, Samo Horvat, Vojko Flis, Nejc Plohl, Izidor Mlakar, 2025, original scientific article Abstract: Aims and Objective: The aim of this study was to explore the feasibility, usability and acceptance of integrating Clinical Decision Support Systems with Socially Assistive Robots into hospital grand rounds. Background: Adopting Clinical Decision Support Systems in healthcare faces challenges such as complexity, poor integration with workflows, and concerns about data privacy and quality. Issues such as too many alerts, confusing errors, and difficulty using the technology in front of patients make adoption challenging and prevent it from fitting into daily workflows. Making Clinical Decision Support System simple, intuitive and user-friendly is essential to enable its use in daily practice to improve patient care and decision-making. Methods: This six-month pilot study had two participant groups, with total of 40 participants: a longitudinal intervention group (n =8) and a single-session evaluation group (n=32). Participants were medical doctors at the University Clinical Center Maribor. The intervention involved implementing a Clinical Decision Support System delivered via a Socially Assistive Robot during hospital grand rounds. We developed a system that employed the HL7 FHIR standard for integrating data from hospital monitors, electronic health records, and patient-reported outcomes into a single dashboard. A Pepper-based SAR provided patient specific recommendations through a voice and SAR tablet enabled interface. Key evaluation metrics were assessed using the System Usability Scale (SUS) and the Unified Theory of Acceptance, Use of Technology (UTAUT2) questionnaire, including Effort Expectancy, Performance Expectancy and open ended questions. The longitudinal group used the system for 6 months and completed the assessments twice, after one week and at the end of the study. The single-session group completed the assessment once, immediately after the experiment. Qualitative data were gathered through open-ended questions. Data analysis included descriptive statistics, paired t-tests, and thematic analysis. Results: System usability was rated highly across both groups, with the longitudinal group reporting consistently excellent scores (M =82.08 at final evaluation) compared to the acceptable scores of the single-session group (M =68.96). Extended exposure improved user engagement, reflected in significant increases in Effort Expectancy and Habit over time. Participants found the system enjoyable to use, and while no significant changes were seen in Performance Expectancy, feedback emphasized its efficiency in saving time and improving access to clinical data, supporting its feasibility and acceptability. Conclusions: This research supports the potential of robotic technologies to transform CDSS into more interactive, efficient, and user-friendly tools for healthcare professionals. The paper also suggests further research directions and technical improvements to maximize the impact of innovative technologies in healthcare. Keywords: clinical decision support systems, clinical decision-making, hospital grand rounds, patient data integration, perceived quality of care, socially assistive robots, usability and familiarity, user experience questionnaire, workload reduction Published in DKUM: 30.05.2025; Views: 0; Downloads: 2
Full text (3,02 MB) |
4. Compression of triangulated solids’ surfaces by Decimating Reconstructable TrianglesBlaž Repnik, Borut Žalik, Krista Rizman Žalik, 2025, original scientific article Abstract: This paper introduces a new programming solution for reducing the size of files needed to store 3D geometric solids, whose surfaces are interpolated by watertight irregular triangle meshes. Unlike other approaches, the software does not introduce any special storage data format, but, instead, utilises popular CAD formats such as STL, OBJ, or PLY. This increases the software’s interoperability significantly. The software comprises an encoder and a decoder. The encoder estimates which facets may be removed in such a way that they can later be reconstructed unambiguously by the decoder. The decoder also ensures that the reconstructed triangles are oriented correctly. The encoder and the decoder are straightforward to understand and to implement. They are asymmetric, making any potential programming error easier to identify and correct. Keywords: data compression, geometric modelling, computer-aided design, boundary representation, topology, watertightness Published in DKUM: 30.05.2025; Views: 0; Downloads: 1
Full text (2,71 MB) |
5. High-performance deployment operational Data analytics of pre-trained multi-label classification architectures with differential-evolution-based hyperparameter optimization (AutoDEHypO)Teo Prica, Aleš Zamuda, 2025, original scientific article Abstract: This article presents a high-performance-computing differential-evolution-based hyperparameter optimization automated workflow (AutoDEHypO), which is deployed on a petascale supercomputer and utilizes multiple GPUs to execute a specialized fitness function for machine learning (ML). The workflow is designed for operational analytics of energy efficiency. In this differential evolution (DE) optimization use case, we analyze how energy efficiently the DE algorithm performs with different DE strategies and ML models. The workflow analysis considers key factors such as DE strategies and automated use case configurations, such as an ML model architecture and dataset, while monitoring both the achieved accuracy and the utilization of computing resources, such as the elapsed time and consumed energy. While the efficiency of a chosen DE strategy is assessed based on a multi-label supervised ML accuracy, operational data about the consumption of resources of individual completed jobs obtained from a Slurm database are reported. To demonstrate the impact on energy efficiency, using our analysis workflow, we visualize the obtained operational data and aggregate them with statistical tests that compare and group the energy efficiency of the DE strategies applied in the ML models. Keywords: high-performance computing, operational data analytics, energy efficiency, machine learning, AutoML, differential avolution, optimization Published in DKUM: 29.05.2025; Views: 0; Downloads: 2
Full text (1,61 MB) |
6. Pull factors and capital inflows : empirical insights from transformative dynamics in Southeast EuropeMehmed Ganić, Nedim Gavranović, 2024, original scientific article Abstract: This study seeks to examine pull factors of capital inflows, offering an
empirical analysis based on a panel study of eleven Southeast
European countries (Albania, Bosnia and Herzegovina, Bulgaria,
Croatia, Greece, Montenegro, North Macedonia, Kosovo, Romania,
Serbia, and Türkiye) over the period of 2004 – 2021. Methodologically,
the study utilizes a fixed effects (FE) regression model with robust
Driscoll-Kraay standard errors to address issues of heteroskedasticity,
autocorrelation, and potential cross-country correlation. The study
finds that several pull factors can be relevant in driving capital inflows
as follows: market size, inflation, financial and trade openness. The
empirical analysis confirms that the forces of trade liberalization,
financial liberalization, market size, real interest rates and inflation
stability are the elements that encourage capital inflows. On the other
hand, the estimated effects of current account balance and real
economic growth are not very convincing. Finally, we stress that more
study is required to fully understand the pull variables' ultimate
macroeconomic implications at the national level. The overall
influence of these positive (or negative) inflows may be moderated by
several characteristics, even if certain countries may be extremely
susceptible to these factors. Keywords: capital inflows, pull factors, southeast Europe, panel data analysis, Driscoll-Kraay standard errors Published in DKUM: 28.05.2025; Views: 0; Downloads: 0
Full text (611,04 KB) This document has many files! More... |
7. Sustainable development goals and higher education : an efficiency analysisMaja Mihaljević Kosor, 2023, original scientific article Abstract: Higher education (HE) is a significant factor in a country’s economic
prosperity and plays a vital role in addressing sustainability issues and
actively promoting sustainable development. While many EU member
countries have well-developed education systems in terms of rankings and
SDGs’ attainment, little is known about the progress of other European
countries. The goal of this research is to estimate the efficiency of higher
education in the attainment of Sustainable Development Goals (SDGs) in
40 European countries. The method used to estimate efficiency is Data
Envelopment Analysis (DEA) with output-orientation and variable returns
to scale approach. In the final model specification, two input variables and
one output variable are used. Results indicate that the average technical
efficiency of the 40 European countries is relatively high and equal to 0.94.
Nine countries emerge as fully efficient in achieving SDG 4 with a coefficient
equal to 1. The four largest higher education systems achieved an aboveaverage efficiency score of 0.97 or higher. Six countries are recognized as
the worst performing. However, more analysis is necessary to examine the
sources of inefficiency in the worst-performing countries. Due to specific data
limitations indicated in this research, it remains a challenge to evaluate the
precise impact of higher education and its contribution to SDGs. Keywords: sustainable development goals (SDGs), higher education, data envelopment analysis, SDG4, European countries Published in DKUM: 28.05.2025; Views: 0; Downloads: 1
Full text (451,92 KB) This document has many files! More... |
8. EU Health Union : legal aspects and digital healthJasmina Cvahte, Andreja Primec, 2024, review article Abstract: Recurrent health crises and outbreaks of infectious diseases of
international concern have created a specific EU public health policy.
Its role was further strengthened by the COVID-19 pandemic, which
triggered the next stage in the integration process of the Member
States: the European Health Union. The research is carried out using a
literature review method (a review of literature and documents in the
field of public health policy and the European Health Union) and a
comparative-legal method for comparing EU legal acts on which the
Health Union is based (endogenous comparison) and the legal acts of
EU and the Republic of Slovenia (exogenous comparison). The findings
are examined using a qualitative content analysis method, which
allows for meaningful aggregation and use of the data under study to
answer the research questions. The European Health Union builds on
and extends existing areas of public health and requires greater
integration between Member States. One of its most essential pillars is
a single information system with a single database to improve the
health of individuals, resilience to cross-border health crises, patient
mobility, and joint research on the most severe diseases. Member
States, including Slovenia, must follow EU public health commitments.
As the Slovenian legislative proposal analysis on the digitalization of
healthcare shows, this challenging task will also require respect for the
institute (safeguards) of other areas of law. Keywords: European Health Union, public health, European Data Space, health information system, regulatory framewor Published in DKUM: 28.05.2025; Views: 0; Downloads: 0
Full text (328,51 KB) This document has many files! More... |
9. The OpenScience Slovenia metadata datasetMladen Borovič, Marko Ferme, Janez Brezovnik, Sandi Majninger, Albin Bregant, Goran Hrovat, Milan Ojsteršek, 2020, other scientific articles Abstract: The OpenScience Slovenia metadata dataset contains metadata entries for Slovenian public domain academic documents which include undergraduate and postgraduate theses, research and professional articles, along with other academic document types. The data within the dataset was collected as a part of the establishment of the Slovenian Open-Access Infrastructure which defined a unified document collection process and cataloguing for universities in Slovenia within the infrastructure repositories. The data was collected from several already established but separate library systems in Slovenia and merged into a single metadata scheme using metadata deduplication and merging techniques. It consists of text and numerical fields, representing attributes that describe documents. These attributes include document titles, keywords, abstracts, typologies, authors, issue years and other identifiers such as URL and UDC. The potential of this dataset lies especially in text mining and text classification tasks and can also be used in development or benchmarking of content-based recommender systems on real-world data. Keywords: metadata, real world data, text data, text mining, text identification, natural language processing Published in DKUM: 22.05.2025; Views: 0; Downloads: 6
Full text (187,50 KB) This document has many files! More... |
10. An algorithm for voxelised solids representation using chain codesBlaž Repnik, Libor Váša, Borut Žalik, 2025, original scientific article Abstract: The paper introduces a new method to describe the surfaces of voxelised solids. It operates in three stages: a hierarchical linked list of chain code sequences is created first; the linked lists are pruned; and, finally, the content of the data structure is stored. The method uses chain codes from either a three- or nine-symbols alphabet. In the first case, two chain code symbols are needed to access the next face, while, in the second case, this is done by one symbol. The pair of chain codes from the three-symbols alphabet, or the individual symbol from the nine-symbols alphabet are considered as tokens. The sets of tokens are, in both cases, extended by two tokens, indicating the beginning and ending of the list. The method processes solids of any shape, including those containing holes, cavities, or multiple components existing in the same voxel space. Edge-connectivity is permitted. The method was compared against the method proposed by Lemus et al., which is designed for solids without holes. Although supporting a broader set of voxelised solids, the proposed method generates sequences of tokens that are, on average, up to 10% shorter. Since the information entropy of the sequences of tokens produced by the proposed method is also smaller, the obtained sequences are more compressible, as confirmed by applying gzip and bzip2 data compressors. Keywords: voxels, chain code, data structures, pruning, information entropy, data compression Published in DKUM: 19.05.2025; Views: 0; Downloads: 1
Full text (11,28 MB) |