This study conducts a comparative analysis of various machine learning and deep learning models for predicting order quantities in supply chain tiers. The models employed include XGBoost, Random Forest, CNN-BiLSTM, Linear Regression, Support Vector Regression (SVR), K-Nearest Neighbors (KNN), Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN), Bidirectional LSTM (BiLSTM), Bidirectional GRU (BiGRU), Conv1D-BiLSTM, Attention-LSTM, Transformer, and LSTM-CNN hybrid models. Experimental results show that the XGBoost, Random Forest, CNN-BiLSTM, and MLP models exhibit superior predictive performance. In particular, the XGBoost model demonstrates the best results across all performance metrics, attributed to its effective learning of complex data patterns and variable interactions. Although the KNN model also shows perfect predictions with zero error values, this indicates a need for further review of data processing procedures or model validation methods. Conversely, the BiLSTM, BiGRU, and Transformer models exhibit relatively lower performance. Models with moderate performance include Linear Regression, RNN, Conv1D-BiLSTM, Attention-LSTM, and the LSTM-CNN hybrid model, all displaying relatively higher errors and lower coefficients of determination (R²). As a result, tree-based models (XGBoost, Random Forest) and certain deep learning models like CNN-BiLSTM are found to be effective for predicting order quantities in supply chain tiers. In contrast, RNN-based models (BiLSTM, BiGRU) and the Transformer show relatively lower predictive power. Based on these results, we suggest that tree-based models and CNN-based deep learning models should be prioritized when selecting predictive models in practical applications.

The northern territories of Russia need high-quality strategic digital changes in the structure of the regional economy. Digitalization and the introduction of digital technologies in the medium term will be able to transform economic relations in the old industrial and raw materials regions of the North, improve the quality of life of local communities. The growth of digital inequality among the regions under study leads to disproportions in their socio-economic development. The purpose of this study is to develop and test a methodology for assessing the level of development of the digital infrastructure of the Russian northern regions, including classification of an indicators system for each level of digital infrastructure, calculation of an integral index and typology of the territories under study. The objects of the study were 13 northern regions of the Russian Federation, the entire territory of which is classified as regions of the Extreme North and equivalent areas. The methodology made it possible to determine the level of technical, technological and personnel readiness of the northern regions for digitalization, to identify regions with the best solutions at each level of digital infrastructure development. The analysis of the results in dynamics helped to assess the effectiveness of regional policy for managing digitalization processes. As a result, the authors came to the conclusion that increasing the competitiveness of northern regions in the era of rapid digitalization is possible through investments in human capital and the creation of a network of scientific and technological clusters. The presented approach to assessing the development of individual levels and elements of digital infrastructure will allow for the diagnosis of priority needs of territories under study in the field of digitalization. The results of the study can form the basis for regional policy in the field of sustainable digital development of Russia.

The National Fitness Program Plan (2021–2025) (hereinafter referred to as the Plan) proposes to perfect the public service system for sports and fitness by 2025, make national sports and fitness more convenient, and advocate providing intelligent services for national fitness campaign. With the development of the Internet era, modern information technologies such as big data, the Internet of Things, and artificial intelligence have been introduced into sports affairs, providing technical support for the optimization of the public service system for sports and fitness. Therefore, in the context of a national fitness campaign, intelligent sports service is an important link for promoting national fitness in various regions. Relevant workers should attach importance to promoting “physical fitness” with “intelligence” in the process of advancing national fitness program, and actively creating intelligent public services for national fitness. Focusing on the integration of modern information technology and sports affairs, with the implementation of the Plan as the research background, the construction of intelligent sports parks as the starting point, this article outlines the construction plan of intelligent sports parks based on the connotation summary of national fitness program and intelligent sports. At the same time, it analyzes the issues that intelligent sports parks need to pay attention to in providing public services for national fitness, and proposes countermeasures for the high-quality development of national fitness services in intelligent sports parks.

Ukrainian Human Resource (HR) practices have multiple difficulties from economic changes combined with digital transformation and workforce instability brought on by the war in 2022. The study examines Ukrainian HR practices between 2015 and 2024, focusing on the digitalization of HR systems, talent development, staff engagement, and hiring strategies. It considers the effects of organizational size and industry type. The study combined interviews with 30 HR professionals and surveyed 150 organizations from different industry groups and sizes. Our data required both quantitative statistical tests and manual content breakdown with codes. Research has shown significant differences between Information Technology (IT) and farming firms, as 89% of IT businesses have integrated artificial intelligence (AI)-powered HR tools. In comparison, only 15% of agricultural companies have adopted them. Small and medium-sized enterprises (SMEs) showed less commitment to digital transformation and European Union (EU) requirements than large enterprises, which adopted these systems at rates of 75% and 88%, respectively. Western Ukraine first established mental health initiatives during the crisis, and Eastern Ukraine moved toward decentralized administration. Digitalization assistance for small businesses, along with EU and local human resources frameworks, should form the basis of our suggestions. This research calls for flexible people management methods to boost the Ukrainian workspace’s ability to recover from shocks.

While the rapid development of artificial intelligence has affected people's daily lives, it has also brought huge challenges to high school mathematics teaching, such as restructuring the classroom teaching structure, transforming the role of teachers, and selecting classroom teaching methods. Based on this, the article explores the application strategies of AI technology in improving knowledge introduction, improving mathematics classroom efficiency and stimulating students' learning interest, with a view to optimizing classroom teaching links, improving students' core discipline quality, and promoting the development of high school mathematics teaching informatization.

Herein, we report a facile preparation of super-hydrophilic sand by coating the sand particles with cross-linked polyacrylamide (PAM) hydrogels for enhanced water absorption and controlled water release aimed at desert agriculture. To prepare the sample, 4 wt% of aqueous PAM solution is mixed with organic cross-linkers of hydroquinone (HQ) and hexamethylenetetramine (HMT) in a 1:1 weight ratio and aqueous potassium chloride (KCl) solution. A specific amount of the above solution is added to the sand, well mixed, and subsequently cured at 150 °C for 8 h. The prepared super-hydrophilic sands were characterized by Fourier-transform infrared spectroscopy (FT-IR) for chemical composition and X-ray diffraction (XRD) for successful polymer coating onto the sand. The water storage for the samples was studied by absorption kinetics at various temperature conditions, and extended water release was studied by water desorption kinetics. The water swelling ratio for the super-hydrophilic sand has reached a maximum of 900% (9 times its weight) at 80 °C within 1 h. The desorption kinetics of the samples showed that the water can be stored for up to a maximum of 3 days. Therefore, super-hydrophilic sand particles were successfully prepared by coating them with PAM hydrogels, which have great potential to be used in sustainable desert agriculture.