101. Analyzing EEG signal with Machine Learning in Python : graduation thesisEvgenija Siljanovska, 2023, diplomsko delo Opis: This thesis presents a comprehensive analysis of EEG data using Python libraries, MNE and machine learning techniques. The thesis focuses on utilizing these tools to extract valuable insights from EEG recordings. Our dataset consists of EEG data in the BrainVision format, acquired during a psychology experiment. The analysis involves preprocessing, filtering, segmentation, and visualization of the EEG data. Additionally, machine learning algorithms are employed to classify and predict patterns within the EEG signals. The findings showcase the effectiveness of Python, MNE, and machine learning in EEG analysis.
Ključne besede: EEG data, MNE, Machine learning, Analyzing
Objavljeno v DKUM: 17.08.2023; Ogledov: 585; Prenosov: 75
 Celotno besedilo (2,66 MB) |
102. Scoping review on the multimodal classification of depression and experimental study on existing multimodal modelsUmut Arioz, Urška Smrke, Nejc Plohl, Izidor Mlakar, 2022, pregledni znanstveni članek Opis: Depression is a prevalent comorbidity in patients with severe physical disorders, such as cancer, stroke, and coronary diseases. Although it can significantly impact the course of the primary disease, the signs of depression are often underestimated and overlooked. The aim of this paper was to review algorithms for the automatic, uniform, and multimodal classification of signs of depression from human conversations and to evaluate their accuracy. For the scoping review, the PRISMA guidelines for scoping reviews were followed. In the scoping review, the search yielded 1095 papers, out of which 20 papers (8.26%) included more than two modalities, and 3 of those papers provided codes. Within the scope of this review, supported vector machine (SVM), random forest (RF), and long short-term memory network (LSTM; with gated and non-gated recurrent units) models, as well as different combinations of features, were identified as the most widely researched techniques. We tested the models using the DAIC-WOZ dataset (original training dataset) and using the SymptomMedia dataset to further assess their reliability and dependency on the nature of the training datasets. The best performance was obtained by the LSTM with gated recurrent units (F1-score of 0.64 for the DAIC-WOZ dataset). However, with a drop to an F1-score of 0.56 for the SymptomMedia dataset, the method also appears to be the most data-dependent.
Ključne besede: multimodal depression classification, scoping review, real-world data, mental health
Objavljeno v DKUM: 11.08.2023; Ogledov: 529; Prenosov: 95
Celotno besedilo (1,43 MB)
Gradivo ima več datotek! Več... |
103. |
104. Development and optimisation of a business intelligence system : master's thesisMark Žohar, 2023, magistrsko delo Opis: In the fast-paced business landscape, unexpected events can pose significant challenges for companies. A lot of times, business intelligence presents a solution for companies in dealing with changes and improving organizational efficiency. As part of this master’s thesis, we have developed a business intelligence system that assists in the planning of new releases of smartphone components. We have simulated data and created reports, which can help the stakeholders to plan and analyze important milestones easily. We have tested two different data structures and showed how star schema is the most beneficial data structure for our BI system. We achieved this by restructuring data into a star schema model at different stages of migration and then compared the size of data and responsiveness of operations.
Ključne besede: business intelligence, star schema, data modeling
Objavljeno v DKUM: 27.06.2023; Ogledov: 414; Prenosov: 62
Celotno besedilo (1,69 MB) |
105. 36th Bled eConference – Digital Economy and Society: The Balancing Act for Digital Innovation in Times of Instability : June 25 – 28, 2023, Bled, Slovenia, Conference Proceedings2023 Opis: The Bled eConference, organised by the University of Maribor, Faculty of Organizational Sciences, has been shaping electronic interactions since 1988. The theme of the 36th conference is "Digital Economy and Society: The Balancing Act for Digital Innovation in Times of Instability". In times of instability, which include political, economic, resource, health, and environmental challenges on the one hand, and technological disruption on the other, it is critical to ensure that digital innovation continues to lead to the right and sustainable solutions that are tailored to the needs of all people, enterprises and society. It is very important to keep in mind the protection of our planet, including fauna and flora. These efforts include adopting appropriate regulatory frameworks, fostering digital literacy and skills development, promoting inclusive access to digital technologies, and addressing the ethical, social and environmental implications of digital transformation. The papers in this conference proceedings address digital transformation of enterprises, artificial intelligence and data science solutions, decision analytics for business and societal challenges, new, digital and data driven business models, digital consumer, digital education, digital health, digital ethics, restructured work and solutions for smart and sustainable cities. We continue to provide an open forum for academia, including students, industry, and policy makers where everyone can contribute to creating a better world.
Ključne besede: digital transformation, digital business, digital technologijes, innovations, digitalization, sustainable development, smart and sustainable cities and societies, digital health, artificial intelligence and data science, digital ethics, digital education, restructured work, digital consumer, smart and sustainable cities
Objavljeno v DKUM: 26.06.2023; Ogledov: 664; Prenosov: 291
Celotno besedilo (17,88 MB)
Gradivo ima več datotek! Več... |
106. FDI and economic growth : a new look from the sectoral perspectiveVladimir Šimić, Lena Malešević-Perović, 2022, izvirni znanstveni članek Opis: This paper investigates the impact of foreign direct investment (FDI) on
economic growth in a sample of 10 ex-socialist European and Asian countries.
While the link between FDI and economic growth has been extensively
investigated in empirical literature, this paper contributes to this literature
by econometrically investigating the effects of FDI inflows in specific sectors
using the panel data estimation techniques. In addition to confirming the
positive impact of total FDI inflows on economic growth, using the WiiW
(the Vienna Institute for International Economic Studies) FDI database,
which disentangles the FDI data across different industries, the paper finds
significant effects of FDI inflows in manufacturing on growth, whereas FDI
inflows in other sectors are only sporadically statistically significant. This
finding serves as a basis for further investigation of the specific subsectors
(NACE 2-letter classification) within manufacturing, and the empirical
investigation finds that not all FDI inflows within manufacturing enhance
economic growth.
Ključne besede: FDI, economic growth, sectoral data, manufacturing
Objavljeno v DKUM: 19.06.2023; Ogledov: 435; Prenosov: 8
 Povezava na datoteko
Gradivo ima več datotek! Več... |
107. Proceedings of the 33rd International Conference on Information Modelling and Knowledge Bases EJC 20232023 Opis: The proceedings of the thirty-third conference EJC 2023 combine the experience and knowledge of the experts working and different research fields of Information modelling, Conceptual modelling, Knowledge and information modelling and discovery, Linguistic modelling, Cross-cultural communication and social computing, Environmental modelling and engineering, and Multimedia data modelling and systems.
Ključne besede: conceptual modelling, knowledge modelling, information modelling, linguistic modelling, cross-cultural communication, social computing, environmental modelling, multimedia data modelling
Objavljeno v DKUM: 02.06.2023; Ogledov: 986; Prenosov: 32
Celotno besedilo (30,93 MB)
Gradivo ima več datotek! Več... |
108. Cyber Security- Training Students and Scholars for the Challenges of Information and Communication Technologies in Research and Studies for Internationalisation : handbook2023 Opis: This handbook is a product of the Erasmus+ Strategic Partnership between the partners Université Polytéchnique Hauts-de-France, Politechnika Poznanska, Brandenburgische Technische Universität Cottbus-Senftenberg and the coordinator University of Maribor. Contentuous contribution were made by staff of University of Maribor and Politechnika Poznanska, evaluations by staff of the other two partners. The handbooks handles virtual learning environments in the international education and research area and exposure of the systems to cybercrime. Besides introductory contributions on legal aspects of challenges in the fields of human rights, European regulations of data security, civil law and criminal law aspects of caber security of virtual learning environments and methodologies of their introduction w8ithin organisations, the handbook also gives useful instructions for elaboration of virtual courses within virtual learning environments relevant for the internationalised research and education not only since the Covid-19 pandemic. The handbook is dedicated to use on the internet within the Moodle system.
Ključne besede: cyber security, virtual learning environment, data security, cyber crime, systems
Objavljeno v DKUM: 13.04.2023; Ogledov: 554; Prenosov: 43
Celotno besedilo (5,72 MB)
Gradivo ima več datotek! Več... |
109. Application of normalization method to fracture toughness testing of welds with pronounced strength heterogeneity : doctoral disertationPrimož Štefane, 2022, doktorska disertacija Opis: This doctoral dissertation presents the results of an extensive fracture testing programme of
welds with pronounced strength heterogeneity. Purpose of this programme was to determine
fracture toughness of heterogeneous welds that contain a midplane crack. Application of
standardized fracture testing methods in heterogeneous welds might lead to overestimation or
underestimation of fracture toughness and consequentially to inaccurate assessment of
structural integrity. Reasons for that are variations in mechanical properties of different
material regions in the weld which have a significant impact on development of deformation at
the crack tip, and consequently on the crack driving force. Experimental procedures in scope
of this research include fabrication of weld sample plates, that were welded with MAG process.
The welds were fabricated using two different electrodes, one with higher and one with lower
mechanical properties, with respect to base material S690QL in order to replicate extreme
variations of mechanical properties in the weldment. Fabricated welds were then characterized
in detail using metallography, three-point bend impact testing, indentation hardness
measurements and tensile testing of flat miniature and round bar standard tensile specimens.
Resistance of welds to stable tearing was investigated by fracture testing of square surface
cracked SE(B) specimens containing a weld midplane notch. J-integral has been estimated from
plastic work, using the normalization data reduction method that is included in standard ASTM
E1820. The advantage of the normalization data reduction method is that no special equipment
or complex testing method is needed to measure ductile crack growth during fracture testing.
The ductile crack growth is determined directly from the load-displacement record, by applying
appropriate calibration function and physical lengths of initial and final cracks that were
measured post-mortem with the nine-point method. Several correction factors had to be
calibrated in order to successfully implement the normalization data reduction method to
fracture testing of welds with pronounced strength heterogeneity. For that reason, parametric
finite element analyses were conducted for several weld configurations. Finite element models
incorporated plane strain conditions in order to provide calibrated factors that comply with
plane strain equations included in ASTM E1820. Additionally, crack tip constraint has been
extensively analysed and correlated with the plastic deformation fields. This clarified altered
deformation behaviour of modelled welds in comparison with the base material and
corresponding effect on fracture toughness. Finally, calibrated factors were applied to
computation of J-integral from data that were measured during fracture testing. J-R resistance
curves were constructed for the tested heterogeneous welds and compared to the ones of the
base material. This directly showed the effect of variations of mechanical properties on the
weld fracture behaviour.
Ključne besede: weld, strength mismatch, fracture, normalization data reduction technique, plastic
correction factors, test fixture, SE(B) specimen, J-R resistance curve
Objavljeno v DKUM: 10.01.2023; Ogledov: 778; Prenosov: 214
Celotno besedilo (20,57 MB) |
110. Open Science Summer School : Maribor, Slovenia, 12.-16. September 20222022, druge monografije in druga zaključena dela Opis: Skills and competencies related to data literacy and data management are indispensable in today’s data-intensive research environment, both in STEM as well as in the humanities and social sciences. In recent years, there has been a significant shift in scientific communication towards open science, and many new approaches, tools and technologies have emerged, enabling a research process that goes beyond the traditional way of research. Open science has the potential to lead researchers to an open, collaborative, transparent, reproducible and, therefore, effective research process.
The University of Maribor Summer School offers PhD students and young researchers an opportunity to learn and develop skills in the fields of research data management, open access to research results, the use of open research infrastructure, supercomputing resources and the European Open Science Cloud. With lectures and workshops led by domestic and international experts, the course will provide a deeper understanding of these topics and insight into valuable practical experience.
Ključne besede: scientific research, open science, open publications, research data
Objavljeno v DKUM: 21.12.2022; Ogledov: 896; Prenosov: 15
Povezava na datoteko
Gradivo ima več datotek! Več... |
