Iskanje po katalogu digitalne knjižnice

Opis: This paper investigates the automatic recognition of depression by integrating acoustic, linguistic and emotional features extracted from clinical interviews in the DAIC-WOZ dataset. A total of six classical machine learning classifiers such as Decision Tree, Random Forest, SVM, Gradient Boosting, AdaBoost and XGBoost were systematically evaluated under different class balancing methods (such as SMOTE, SMOTETomek and Random Undersampling) and feature selection strategies. The best model, a decision tree classifier with SMOTE-based balancing and a feature selection technique, achieved a weighted F1 score and accuracy of 0.78 with only eight selected features. These features included all three modalities, demonstrating the added benefit of a multimodal approach. The results suggest that even relatively simple models, when supported by careful preprocessing and dimensionality reduction, can provide accurate and interpretable predictions. This work emphasizes the importance of feature engineering and balancing techniques in clinical machine learning tasks and lays the foundation for future research on scalable and explainable depression detection systems.
Ključne besede: Depression, Classification, Machine Learning, Data Balancing, Feature Selection
Objavljeno v DKUM: 03.11.2025; Ogledov: 0; Prenosov: 0
Celotno besedilo (4,28 MB)

Opis: 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.
Ključne besede: resonant grounded networks, ground-fault relay, high-impedance faults, ensemble-based learning, optimal feature selection
Objavljeno v DKUM: 25.07.2025; Ogledov: 0; Prenosov: 3
Celotno besedilo (4,86 MB)

Opis: Background: The selection of the right Incoterms is crucial for minimising risks and costs in international trade. This paper aims to develop a model that identifies the key factors influencing the selection of Incoterms. The main contribution of the research is the creation of a new statistical modelling process that effectively identifies the variables impacting trade costs and risk. The study uses import and export data from non-EU countries in the context of a Slovenian case study. Methods: A novel model selection mechanism is developed, combining the logistic regression (logit) modelling with Monte Carlo simulations to identify influential factors in Incoterms selection. This mechanism incorporates heuristic techniques, which guide a sequential process of gradually searching through logit model candidates to determine the bestfit model for both import and export scenarios. Results: The application of the new logit modelling procedure reveals that the delivery location is the most significant factor affecting Incoterms selection. Additionally, the value of goods and the type of transport (containerised vs. noncontainerised) also have a considerable influence. For imports, the mass of goods is found to be a significant factor as well. Conclusions: The research results offer valuable insights for companies formulating their international business strategies. By selecting the appropriate Incoterm, companies can reduce transportation risks and costs. Managing costs and risks is especially important for higher-value goods. The research finds that, for lower-value goods, sellers often take on associated costs and risks. This pattern is particularly notable with imports, where sellers are more likely to assume responsibility for lighter-weight goods.
Ključne besede: international trade, Incoterms, Incoterms selection, logit regression, Monte Carlo simulation
Objavljeno v DKUM: 22.07.2025; Ogledov: 0; Prenosov: 0
Celotno besedilo (1,41 MB)
Gradivo ima več datotek! Več...

Opis: Purpose of the article –The purpose of this article is to explore the correlation between generational values and tourism product selection. It addresses the scientific problem of how socio-historical experiences influence consumer behaviour in the tourism sector. The goal is to identify distinct patterns among Baby Boomers, Generation X, and Generation Y that affect travel preferences, including price sensitivity, trip duration, travel style, and desired levels of autonomy. Research methodology –A two-step research methodology was applied. In the first step, a codebook was created based on the catalogue of trips from a cooperating travel agency. Trips were categorised by seven characteristics such as type, length, and autonomy. In the second step, a dataset of 9605 travellers was analysed using SPSS 24. Generational cohorts were determined based on age, and correlation and cross-tabulation analyses were conducted to uncover significant patterns.
Ključne besede: generational values, tourism product selection, consumer behaviour, generational cohorts, travel preferences
Objavljeno v DKUM: 17.06.2025; Ogledov: 0; Prenosov: 30
Celotno besedilo (435,33 KB)
Gradivo ima več datotek! Več...

Opis: When companies make strategic decisions, responsible persons must take into account as many influential parameters as possible so that the solution to the given problem is the most optimal, that is, they make the most appropriate decision. Multi-criteria decision-making (MCDM) can find a very large application for solving such very complex and important issues, where it is of particular importance that the company makes the most appropriate decision. Making the optimal decision for a given problem directly affects the financial performance of a given company. In this paper, the VIKOR method will be applied, which until now has not been used to solve problems related to the choice of warehouse location, but is very often and very successfully used to solve various complex problems when applying multi-criteria decision making (MCDM). The purpose of this paper is to show that the VIKOR method can be successfully applied to select the optimal warehouse location for a company that has subsidiaries in multiple locations.
Ključne besede: selection, location, warehouse, multi-criteria decision making, VIKOR method
Objavljeno v DKUM: 17.12.2024; Ogledov: 0; Prenosov: 235
Celotno besedilo (448,22 KB)
Gradivo ima več datotek! Več...