Oleg Yarema

The purpose of this study was to investigate how digital competencies and digital human capital contribute to sustainable economic development by influencing productivity, competitiveness, innovation capacity, labour-market transformation, and long-term economic resilience. The materials consisted of open data and analytical reports published by European and international institutions, including sources on digital skills, artificial intelligence adoption, labour-market transformation, and digital development. The methodological basis of the study included comparative analysis, synthesis of institutional and statistical data, structural analysis of labour-market trends, calculation of digital-skills gaps and normalised indicators, scenario analysis, and a strengths, weaknesses, opportunities, and threats matrix. The results showed that digital competencies affect sustainable development through productivity growth, innovation capacity, economic inclusion, labour-market adaptability, and institutional resilience. Particular attention was devoted to evaluating Ukraine’s digital transformation in comparison with the European Union and to identifying strategic policy priorities capable of strengthening digital human capital in the context of European integration and post-war economic recovery. The comparative analysis indicated that Ukraine is close to the European Union average in basic digital skills but still faces significant risks related to uneven access, migration of skilled workers, insufficient artificial intelligence literacy, and outdated educational programmes. The extended quantitative assessment showed that, if the 2021-2025 trend continues, Ukraine may reach only about 64.4 percent of adults with at least basic digital skills by 2030, remaining 15.6 percentage points below the recommended benchmark. The article argued that Ukraine’s digital human-capital strategy should combine lifelong learning, data literacy, artificial intelligence literacy, science, technology, engineering, and mathematics education, digital inclusion, and smart regional policy. The practical value of the study lies in the proposed framework for connecting digital competencies with productivity, competitiveness, and recovery policy

High-dimensional gene expression data pose substantial challenges for machine-learning–based diagnostic modelling due to extreme dimensionality, noise, heterogeneous measurement conditions, and limited sample sizes. These factors strongly affect classifier behaviour and necessitate systematic comparative evaluation under controlled experimental conditions. This study presents a comprehensive assessment of widely used machine-learning classifiers applied to high-dimensional transcriptomic data, with a focus on robustness, stability, and suitability for gene expression–driven decision-support systems. The analysis covers representative model families, including tree-based and ensemble methods, margin-based classifiers, and regularized linear models. All classifiers were evaluated within a unified analytical pipeline incorporating normalization, variance-based filtering, stratified data partitioning, Bayesian hyperparameter optimization, and repeated validation. Performance was assessed using accuracy, precision, recall, F1-score, and balanced accuracy to account for class imbalance and heterogeneous sample structures. Experimental results reveal pronounced differences in classifier performance under high-dimensional conditions. Ensemble-based approaches, particularly Random Forest and gradient boosting models, consistently achieved higher stability and balanced accuracy across heterogeneous datasets. Regularized linear models with elastic-net penalties performed competitively after dimensionality reduction, while support vector machines demonstrated high accuracy but required careful kernel and regularization tuning. In contrast, probabilistic classifiers exhibited increased sensitivity to noise and distributional heterogeneity. Overall, the findings emphasize the importance of ensemble learning and regularization strategies for reliable classification of transcriptomic data and provide a reproducible methodological framework for information-technology–driven diagnostic modelling based on gene expression profiles.

The article explores the possibilities of using Big Data and artificial intelligence technologies to forecast demand for postal and logistics services in Ukraine in conditions of increased turbulence caused by military operations, instability of market processes, and dynamic changes in demand. Key data sources are considered, as well as methods for their processing and analysis to build more accurate and adaptive predictive models. The study paid particular attention to the analysis of modern data mining tools, in particular clustering using the DBSCAN method and time series forecasting using Long Short-Term Memory (LSTM), which are currently the most effective for time series forecasting. A comparative analysis of traditional forecasting methods and models built using artificial intelligence and Big Data technologies was conducted to identify the forecasting method with the lowest percentage of forecast error. The study is based on an analysis of modern scientific publications on the topics of Big Data, machine learning, and logistics.

Ship acoustic signal classification is essential for vessel identification, underwater navigation, and maritime security. Traditional methods struggle with the non-stationary nature and noise of ship acoustic signals, reducing classification accuracy. To address these challenges, we propose an automated pipeline that integrates Empirical Mode Decomposition (EMD), adaptive wavelet filtering, feature selection, and a Bayesian-optimized Random Forest classifier. The framework begins with EMD-based decomposition, where the most informative Intrinsic Mode Functions (IMFs) are selected using Signal-to-Noise Ratio (SNR) analysis. Wavelet filtering is applied to reduce noise, with optimal wavelet parameters determined via SNR and Stein’s Unbiased Risk Estimate (SURE) criteria. Features extracted from statistical, frequency domain (FFT), and time–frequency (wavelet) metrics are ranked, and the top 11 most important features are selected for classification. A Bayesian-optimized Random Forest classifier is trained using the extracted features, ensuring optimal hyperparameter selection and reducing computational complexity. The classification results are further enhanced using a majority voting strategy, improving the accuracy of the final object identification. The proposed approach demonstrates high accuracy, improved noise suppression, and robust classification performance. The methodology is scalable, computationally efficient, and suitable for real-time maritime applications.

У статті представлено розробку та застосування гібридної моделі для оцінювання ефективності метрик близькості у високорозмірних даних експресії генів, яка інтегрує методи інтелектуального аналізу даних та машинного навчання в єдиній комплексній структурі. Основна увага приділяється порівняльному аналізу кореляційної відстані, метрик, заснованих на взаємній інформації, та метрики Васерштейна, з метою визначення їхньої ефективності для завдань кластеризації та класифікації. Отримані результати демонструють, що кореляційна відстань і метрика Васерштейна забезпечують високу точність та стабільність, що робить їх придатними для інтеграції з діагностичними системами. Для підвищення надійності класифікації реалізовано стекінг-модель, яка компенсує можливі помилки кластеризації та забезпечує стабільну продуктивність незалежно від використаної метрики та структури кластерів. Запропонований конвеєр обробки даних забезпечує автоматизований, стандартизований і масштабований аналіз великих масивів даних експресії генів, що узгоджується із принципами персоналізованої медицини. Завдяки можливості ранньої діагностики захворювань та підтримці розроблення індивідуальних стратегій лікування отримані результати мають велике значення для вдосконалення сучасних діагностичних систем у межах концепції прецизійної медицини. Застосування гібридних підходів дозволяє ефективно поєднувати переваги різних методів оцінювання подібності та класифікації, що сприяє підвищенню точності прогнозування на основі даних експресії генів. Запропонований методологічний підхід може бути використаний у біоінформатиці для аналізу складних біологічних систем, оптимізації процесів обробки генетичних даних і розроблення нових алгоритмічних рішень у сфері медичної діагностики. Такий підхід сприяє адаптації сучасних інформаційних технологій для виявлення біомаркерів захворювань, що забезпечує інтеграцію отриманих результатів у клінічну практику та підтримує розвиток персоналізованих методів лікування.

The growing volume and complexity of gene expression data necessitate biologically meaningful and statistically robust methods for feature selection to enhance the effectiveness of disease diagnosis systems. The present study addresses this challenge by proposing a pipeline that integrates RNA-seq data preprocessing, differential gene expression analysis, Gene Ontology (GO) enrichment, and ensemble-based machine learning. The pipeline employs the non-parametric Kruskal–Wallis test to identify differentially expressed genes, followed by dual enrichment analysis using both Fisher’s exact test and the Kolmogorov–Smirnov test across three GO categories: Biological Process (BP), Molecular Function (MF), and Cellular Component (CC). Genes associated with GO terms found significant by both tests were used to construct multiple gene subsets, including subsets based on individual categories, their union, and their intersection. Classification experiments using a random forest model, validated via 5-fold cross-validation, demonstrated that gene subsets derived from the CC category and the union of all categories achieved the highest accuracy and weighted F1-scores, exceeding 0.97 across 14 cancer types. In contrast, subsets derived from BP, MF, and especially their intersection exhibited lower performance. These results confirm the discriminative power of spatially localized gene annotations and underscore the value of integrating statistical and functional information into gene selection. The proposed approach improves the reliability of biomarker discovery and supports downstream analyses such as clustering and biclustering, providing a strong foundation for developing precise diagnostic tools in personalized medicine.