1. 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) |
2. LLM in the loop: a framework for contextualizing counterfactual segment perturbations in point cloudsVeljka Kočić, Niko Lukač, Dzemail Rozajac, Stefan Schweng, Christoph Gollob, Arne Nothdurft, Karl Stampfer, Javier Del Ser, Andreas Holzinger, 2025, original scientific article Abstract: Point Cloud Data analysis has seen a major leap forward with the introduction of PointNet algorithms, revolutionizing how we process 3D environments. Yet, despite these advancements, key challenges remain, particularly in optimizing segment perturbations to influence model outcomes in a controlled and meaningful way. Traditional methods struggle to generate realistic and contextually appropriate perturbations, limiting their effectiveness in critical applications like autonomous systems and urban planning. This paper takes a bold step by integrating Large Language Models into the counterfactual reasoning process, unlocking a new level of automation and intelligence in segment perturbation. Our approach begins with semantic segmentation, after which LLMs intelligently select optimal replacement segments based on features such as class label, color, area, and height. By leveraging the reasoning capabilities of LLMs, we generate perturbations that are not only computationally efficient but also semantically meaningful. The proposed framework undergoes rigorous evaluation, combining human inspection of LLM-generated suggestions with quantitative analysis of semantic classification model performance across different LLM variants. By bridging the gap between geometric transformations and high-level semantic reasoning, this research redefines how we approach perturbation generation in Point Cloud Data analysis. The results pave the way for more interpretable, adaptable, and intelligent AI-driven solutions, bringing us closer to realworld applications where explainability and robustness are paramount.
Keywords: explainable AI, point cloud data, counterfactual reasoning, LiDAR, 3D point cloud data, interpretability, human-centered AI, large language models, K-nearest neighbors
Published in DKUM: 19.05.2025; Views: 0; Downloads: 2
Full text (7,24 MB) |
3. Software self-hosting : a systematic review of quantitative research☆Luka Hrgarek, Lili Nemec Zlatolas, 2025, original scientific article Abstract: In an era marked by heightened concerns surrounding personal privacy and data security, software self-hosting has gained significance as a means for individuals and organizations to reclaim control over their digital assets. This systematic review aims to identify relevant research gaps in the quantitative analysis of self-hosting, primarily focusing on studies employing Structural Equation Modeling (SEM) and regression techniques. Employing a refined version of the Systematic Mapping Process, we analyzed 49 quantitative research papers whose concepts were grouped into 12 substantive groups. The findings reveal a predominant concentration on constructs related to the Technology Acceptance Model (TAM), with limited exploration of self-hosting specifically, overshadowed by an emphasis on cloud computing, the Internet of Things (IoT), and privacy aspects. Our review provides a comprehensive overview of the existing literature and highlights the need for more focused research on self-hosting itself. This systematic review serves as a foundational resource for researchers and practitioners aimed at advancing the discourse on self-hosting.
Keywords: self-hosting, social networking sites, privacy, quantitative, data sovereignty, technology acceptance model, systematic literature review
Published in DKUM: 23.04.2025; Views: 0; Downloads: 8
Full text (1,98 MB) |
4. Data management and academic integrityMilan Ojsteršek, 2024, other educational material Abstract: Sensitive data requires careful consideration and adherence to best practices to ensure its confidentiality, integrity, and availability. Essential steps in handling sensitive data are identification and classification of sensitive data, implementation of data access control, encryption of sensitive data, secure storage and transmission, implementation of data breach response plan, backup and monitoring usage of data, complying with regulation, and disposing of data securely. Misconduct in handling sensitive data can compromise data confidentiality, integrity, and availability. These include data breaches (unauthorised access or disclosure, theft, insider threats, falsification, fabrication, imputation, and amputation of data), failure to comply with data protection regulations, inadequate data security practices, improper retention and disposal of data, and failure to report data breaches and incidents. In this presentation Milan Ojsteršek presents how to manage sensitive data, desensitise it, and which are the most common breaches in handling sensitive data incidents. This presentation was given at the 4th ENAI Academic Integrity Summer School 2024, 16th – 21th September 2024, University of Konstanz, Germany.
Keywords: open science, metadata, research data management, sensitive data, academic integrity, data management ethics, research misconduct, licensing of open data, FAIR, Slovenian open access infrastructure
Published in DKUM: 18.04.2025; Views: 0; Downloads: 2
 Full text (16,66 MB) |
5. Prepoznavanje rastlin in njihovih bolezni z mobilno aplikacijoRok Trunkelj, 2025, undergraduate thesis Abstract: Raziskava obravnava prepoznavanje izbranih rastlin in njihovih bolezni s pomočjo
mobilne aplikacije. Na kratko so predstavljena uporabljena orodja: Orange Data mining,
Android Studio, MS Visio, Figma, Flask, Nginx in Gunicorn. Arhitektura rešitve obsega
virtualno okolje v oblaku s strežniki za dostop do modelov za klasifikacijo rastlin in
njihovih bolezni in aplikacijo Android. Opisan je postopek izdelave modelov strojnega
učenja, ki so bili preneseni na strežnik. V nalogi so prikazani pomembni deli kode in
podana razlaga vseh aspektov delovanja aplikacije.
Keywords: umetna inteligenca, razvoj aplikacije, analiza podatkov, podatkovno
rudarjenje, Orange Data Mining.
Published in DKUM: 09.04.2025; Views: 0; Downloads: 9
Full text (3,07 MB) |
6. Real-life application of a wearable device towards injury prevention in tennis : a single-case studyIztok Kramberger, Aleš Filipčič, Aleš Germič, Marko Kos, 2022, original scientific article Abstract: The purpose of this article is to present the use of a previously validated wearable sensor
device, Armbeep, in a real-life application, to enhance a tennis player’s training by monitoring and
analysis of the time, physiological, movement, and tennis-specific workload and recovery indicators,
based on fused sensor data acquired by the wearable sensor—a miniature wearable sensor device,
designed to be worn on a wrist, that can detect and record movement and biometric information,
where the basic signal processing is performed directly on the device, while the more complex
signal analysis is performed in the cloud. The inertial measurements and pulse-rate detection of
the wearable device were validated previously, showing acceptability for monitoring workload and
recovery during tennis practice and matches. This study is one of the first attempts to monitor the
daily workload and recovery of tennis players under real conditions. Based on these data, we can
instruct the coach and the player to adjust the daily workload. This optimizes the level of an athlete’s
training load, increases the effectiveness of training, enables an individual approach, and reduces the
possibility of overuse or injuries. This study is a practical example of the use of modern technology
in the return of injured athletes to normal training and competition. This information will help tennis
coaches and players to objectify their workloads during training and competitions, as this is usually
only an intuitive assessment.
Keywords: tennis, training, data-based coaching, shot recognition, wearable device, workload, recovery
Published in DKUM: 31.03.2025; Views: 0; Downloads: 3
Full text (4,62 MB)
This document has many files! More... |
7. On the use of morpho-syntactic description tags in neural machine translation with small and large training corporaGregor Donaj, Mirjam Sepesy Maučec, 2022, original scientific article Abstract: With the transition to neural architectures, machine translation achieves very good quality for several resource-rich languages. However, the results are still much worse for languages
with complex morphology, especially if they are low-resource languages. This paper reports the
results of a systematic analysis of adding morphological information into neural machine translation
system training. Translation systems presented and compared in this research exploit morphological
information from corpora in different formats. Some formats join semantic and grammatical information and others separate these two types of information. Semantic information is modeled using
lemmas and grammatical information using Morpho-Syntactic Description (MSD) tags. Experiments
were performed on corpora of different sizes for the English–Slovene language pair. The conclusions
were drawn for a domain-specific translation system and for a translation system for the general
domain. With MSD tags, we improved the performance by up to 1.40 and 1.68 BLEU points in the
two translation directions. We found that systems with training corpora in different formats improve
the performance differently depending on the translation direction and corpora size.
Keywords: neural machine translation, POS tags, MSD tags, inflected language, data sparsity, corpora size
Published in DKUM: 28.03.2025; Views: 0; Downloads: 5
Full text (448,16 KB)
This document has many files! More... |
8. Evolution of domain-specific modeling language: an example of an industrial case study on an RT-sequencerTomaž Kos, Marjan Mernik, Tomaž Kosar, 2022, original scientific article Abstract: Model-driven engineering is a well-established software development methodology that
uses models to develop applications where the end-users with visual elements model abstractions
from a specific domain. These models are based on domain-specific modeling language (DSML),
which is particular to the problem domain. During DSML use, new ideas emerge and DSMLs evolve.
However, reports on DSML evolution are rare. This study presents a new DSML called RT-Sequencer
that evolved from our DSML Sequencer to support, in addition to the Data Acquisition domain,
also a new domain—Real-Time Control (RTC) systems. The process of defining models with a new
language RT-Sequencer has changed in a way that new end-users were introduced—advanced endusers, which use general-purpose language (GPL) and advanced programming concepts to define
modeling environments for the RT-Sequencer end-users. More specifically, an industrial experience
with the RT-Sequencer is presented, where DSML was opened for extension so that a GPL code
could be inserted into the model to create new visual blocks for the end-user, and the possibility to
adapt and optimize the execution code for a particular task. Our experience shows the specific case
of DSML evolution supporting another problem domain, and the implementation effort needed to
extend domain-specific modeling language with GPL support.
Keywords: model-driven engineering, domain-specific modeling languages, measurement systems, Real-Time Control systems, data acquisition, language evolution, experience report
Published in DKUM: 27.03.2025; Views: 0; Downloads: 2
Full text (1,70 MB)
This document has many files! More... |
9. Predicting corn moisture content in continuous drying systems using LSTM neural networksMarko Simonič, Mirko Ficko, Simon Klančnik, 2025, original scientific article Abstract: As we move toward Agriculture 4.0, there is increasing attention and pressure on the productivity of food production and processing. Optimizing efficiency in critical food processes such as corn drying is essential for long-term storage and economic viability. By using innovative technologies such as machine learning, neural networks, and LSTM modeling, a predictive model was implemented for past data that include various drying parameters and weather conditions. As the data collection of 3826 samples was not originally intended as a dataset for predictive models, various imputation techniques were used to ensure integrity. The model was implemented on the imputed data using a multilayer neural network consisting of an LSTM layer and three dense layers. Its performance was evaluated using four objective metrics and achieved an RMSE of 0.645, an MSE of 0.416, an MAE of 0.352, and a MAPE of 2.555, demonstrating high predictive accuracy. Based on the results and visualization, it was concluded that the proposed model could be a useful tool for predicting the moisture content at the outlets of continuous drying systems. The research results contribute to the further development of sustainable continuous drying techniques and demonstrate the potential of a data-driven approach to improve process efficiency. This method focuses on reducing energy consumption, improving product quality, and increasing the economic profitability of food processing
Keywords: drying, moisture prediction, big data, artificial intelligence, LSTM
Published in DKUM: 21.03.2025; Views: 0; Downloads: 11
Full text (2,99 MB)
This document has many files! More... |
10. The analysis of the effects of a fare free public transport travel demand based on e-ticketingDanijel Hojski, David Hazemali, Marjan Lep, 2022, original scientific article Abstract: The traditional approach in public transport planning was to collect travel demand data for a more extended period and compose timetables to serve this demand. There are two significant identifiable issues. In the rural areas and off-peak hours, public transport operators provide much more capacities than needed. On the other hand, more capacities than scheduled are needed on certain lines at certain departures on some sporadically occurring occasions. The problem is how to react to short-term changes (daily) triggered by exceptional circumstances and events and midterm changes (weekly, monthly basis) in travel demand. We can trigger changes in travel demand chiefly by introducing a desirable (almost for free) tariff system applied to specific populations. No long-term travel response data exists for this kind of intervention, but an immediate response in public transport supply is needed. In Slovenia, public transport for free for the whole population over 65 years was introduced. With the modern ticketing system, which was designed to be as simple as possible for users (that means "check-in only" at the moment of boarding), the research task was to analyze the travel behavior of the retired population, faced with a new attractive option to travel, based on data of purchased tickets and their afterward validation, for better mid-and long-term planning. Our study finds that ITS technology (in this case, e-ticketing system) can satisfactorily solve the discussed planning and management task.
Keywords: fare-free public transport, smart card data collecting, population mobility, travel demand
Published in DKUM: 13.03.2025; Views: 0; Downloads: 1
Full text (2,11 MB)
This document has many files! More... |
