Iskanje po repozitoriju

Leveraging machine learning to predict academic success : a literature review and bibliometric trend analysis
Bor Bregant, Andreja Istenič, 2026, pregledni znanstveni članek

Opis: This study synthesizes research on the use of machine learning (ML) to predict academic success through a three-scope approach. First, a systematic review of 33 review studies from Web of Science (WOS), Scopus, and Taylor & Francis identifies prevailing methods, levels of prediction, and recurrent gaps in the literature. The synthesis of 33 review studies reveals a consolidation around supervised learning approaches and prediction-focused frameworks, with recurring gaps related to real world implementation and pedagogical integration. Second, a qualitative synthesis of the 10 most cited WOS articles provides in-depth insight into influential algorithmic strategies, data sources, and emerging emphases on explainability, human-centered modeling, and early intervention. Third, a bibliometric analysis of 520 WOS articles maps publication trends, country-level productivity, keyword co-occurrence, citation patterns, and international collaboration networks. Results indicate growing publication activity between 2018 and 2025, with research thematically clustering around methodological foundations in educational data mining, model development and prediction strategies, and contextual applications primarily in higher education, with limited chronological differentiation across themes. China leads in publication volume, while major collaboration hubs are concentrated in Australia, India, and Saudi Arabia. Persistent challenges include inconsistent terminology, data heterogeneity, and limited real-world deployment of ML models.
Ključne besede: data mining, machine learning, academic success, education, systematic literature review, bibliometric analysis
Objavljeno v RUP: 08.09.2026; Ogledov: 77; Prenosov: 1
Celotno besedilo (1,91 MB)
Gradivo ima več datotek! Več...

Escape rooms as an active learning didactic strategy : an analysis of motivation, knowledge, and the learning experience
Andreja Špernjak, Nika Židan, 2026, samostojni znanstveni sestavek ali poglavje v monografski publikaciji

Opis: The study investigates escape rooms as a didactic strategy that integrates active learning, collaboration, and game-based motivation in science education. Two versions of an educational escape room on the digestive system were implemented with eighth-grade students: one physical and one virtual. The research employed a comparative descriptive design with 77 participants, using a questionnaire to assess motivation, perceived knowledge, and engagement. Results indicate that both versions of the escape room enhanced students' motivation and learning satisfaction, with no statistically significant differences between the modalities. The physical version fostered more teamwork and sensory engagement, while the virtual version offered flexibility and digital interactivity. The findings confirm that escape rooms - whether physical or virtual - are effective tools for enhancing students' motivation, cooperation, and understanding of biological concepts. The study also emphasises the importance of teacher facilitation, ethical design, and structured reflection as essential components of the didactic framework.
Ključne besede: student-centred teaching, escape rooms, collaborative learning, learning motivation
Objavljeno v RUP: 29.07.2026; Ogledov: 333; Prenosov: 5
Celotno besedilo (123,59 KB)
Gradivo ima več datotek! Več...

Examining sound animation of visual artworks and its connection to the development of rhythmic sensitivity and perception in preschool children
Eda Birsa, Nika Gros, Barbara Kopačin, 2026, izvirni znanstveni članek

Opis: The aim of this study was to examine the relationship between sound animation of visual artworks and the development of rhythmic sensitivity and perception in preschool children. The study employed a qualitative action research design conducted in a kindergarten setting in Slovenia and involved 36 preschool children. Data were collected through observation, interviews, and analysis of children’s visual and musical products, and were analysed using thematic analysis. The results indicate that multisensory and interdisciplinary activities contribute to the development of rhythmic perception, creative expression, and the transfer of rhythm between visual and auditory modalities. The findings emphasise the importance of integrating music and visual arts in early childhood education.
Ključne besede: rhythmic sensitivity, rhythmic perception, multisensory learning, sound animation of visual artworks, interdisciplinary integration
Objavljeno v RUP: 30.06.2026; Ogledov: 436; Prenosov: 20
Celotno besedilo (477,22 KB)
Gradivo ima več datotek! Več...

Efficient material selection for training occluded mmWave radar-based gesture recognisers
Nuwan Attygalle, Luis A. Leiva, Matjaž Kljun, Klen Čopič Pucihar, 2026, izvirni znanstveni članek

Opis: Radar-based gesture recognition has emerged as a promising approach for unobtrusive interaction. Unlike camera-based systems, radar sensors can detect gestures through opaque materials, enabling seamless embedding into various everyday objects. However, it remains unclear how to train models efficiently for robust gesture recognition through diverse materials. To investigate this, we collected a dataset of 17,520 gesture recordings performed through 73 everyday materials. By comparing several material-sampling and data-augmentation strategies, we found that a small carefully selected representative subset of training materials was sufficient to match performance of a classifier trained on the full material dataset. Our results showed that the models trained on 14 quota-sampled materials achieved accuracies of 95.8% and 91.2%, comparable to training on all 73 materials (96.8% and 91.6%) and significantly better than training without material data (66.8% and 65.8%). Among the evaluated sampling approaches, Quota sampling also provided the best overall trade-off between performance and practicality. In contrast, classifiers trained on augmented data performed worse than those trained on actual material-specific data. Taken together, these findings indicate that, for the tested sensor, gesture set, and material collection, carefully selected real-material data offer a practical route to reducing material-specific data collection in radar-based gesture recognition while preserving generalisation. Code, models, and data are available in the public repository, with additional details provided in the supplementary materials: https://gitlab.com/hicuplab/seeing-through.
Ključne besede: deep learning, training optimization for neural networks, millimetre-wave radar, gesture recognition, material selection, sampling, synthetic noise
Objavljeno v RUP: 19.06.2026; Ogledov: 422; Prenosov: 12
Celotno besedilo (15,56 MB)
Gradivo ima več datotek! Več...

PN-SCA : a high generalization and fast profiled SCA based on prototypical networks
Yu Ou, Yongzhuang Wei, Changhai Ou, Enes Pašalić, 2026, izvirni znanstveni članek

Opis: Recently, there has been a growing interest in scenarios that employ a few (imbalanced) power traces for deep learning based side-channel analysis (DL-SCA). Finding a lightweight DL-SCA architecture that is more generalizable and faster to learn is of great importance when handling such situations. In this work, an initial analysis is conducted on the capability of learning and the susceptibility to attacks of prevalent DL-SCA models, focusing on cases that culminate in an unsuccessful attack. Subsequently, a meta-learning technique, known as prototypical networks, is delineated for the construction of a lightweight deep learning framework. In contrast to conventional DL-SCA models, the proposed architecture, designated as PN-SCA, does not predict the probability of a power trace belonging to an intermediate value in classification. Instead, it facilitates the learning of an encoder capable of embedding power consumption data within a latent space, while also establishing templates, or prototypes, for diverse categories. Moreover, we developed a metric that is specifically intended for the selection of hyperparameters due to the unique training phase of PN-SCA. Finally, four distinct scenarios are established with few (including one imbalanced) power traces to evaluate the correctness of our architecture. The results clearly illustrate the advantages of our PN-SCA in terms of generalization, reduced training costs (with a decrease in profiling traces by over 50%), and significantly enhanced attack effect (with a reduction of power traces requirements by over 90%), thus demonstrating notable improvements over existing methodologies
Ključne besede: side-channel analysis, deep learning based SCA, prototypical networks, few-shot learning
Objavljeno v RUP: 20.05.2026; Ogledov: 489; Prenosov: 19
Celotno besedilo (3,78 MB)
Gradivo ima več datotek! Več...