Home – Home – Jornals – Scientific journal “Vestnik NSUEM” 2026 number 2

This study investigates the effectiveness of single and ensemble classifiers for financial fraud detection based on account-to-account money transfer data. We considered Recurrent Neural Networks, Convolutional Neural Networks, Autoencoders, as well as ensemble methods such as Random Forest, XGBoost, and stacking; simple voting was additionally applied to the neural network models. Experiments were conducted using a large-scale, real-world labeled dataset of transaction records. Classification performance was evaluated using Precision, Recall, F1-score, and ROC-AUC metrics. Computational efficiency was assessed based on training time, inference time, memory usage, and average CPU load. The experimental results demonstrated that the stacking ensemble method achieves the highest accuracy, although it is the most resource-intensive. Meanwhile, the accuracy of more efficient boosting and bagging methods is slightly lower but remains comparable to stacking.

The article proposes an algorithm for identifying a model for organizational systems with distributed decision-making centers in conditions of bounded rationality. The method includes the definition of the object and subjects of management, the formation of a system of attributes and the vector formalization of goal functions, taking into account priorities. The approach provides an opportunity for a formal description of conflicts of interest and coordination of managerial influences in a decentralized management environment. The applicability of the approach is shown by the example of the control system of a shipbuilding enterprise.

An important trend in the development of the global economy at the beginning of the 21st century was the increasing role of digital products in the management of production processes. In the modern world, digital transformation is one of the main factors of harmonious economic growth. The mining industry is also not left out of the digital transformation, although the pace of implementation of information and communication technologies should be recognized as not high. The engine of digitalization of the industry can be the exploration sector, the main product of which is information about the deposit, based on the processing and analysis of a huge amount of data presented in various formats. Their processing requires the creation and implementation of specialized software products that collect, store, format, and analyze information. When making a management decision on the use of digital tools in production activities, taking into account the significant initial costs of implementation, it is necessary to evaluate not only the quality characteristics of the product, but also the economic efficiency of its use. The article presents an analysis of the development of digital technologies in the mining industry identifies the reasons for the lag of digitalization in the mineral resource sector compared to other sectors of the economy. The necessity of applying an assessment of economic efficiency in making managerial decisions when implementing digital products at a mining enterprise is substantiated. A methodology for evaluating the economic effectiveness of software products has been developed, which consists of an algorithm and calculated indicators. The technique has been successfully tested at the mining enterprises of Polymetal JSC and has shown its practical significance.

This paper addresses the problem of detecting cognitive fatigue in online learning environments using eye-tracking data. The proposed approach combines statistical analysis (Pearson correlation and one-way ANOVA) with machine learning classification. Based on a publicly available dataset containing 2,784 observations with 41 eye-tracking parameters (including pupil diameter, fixation duration, time to first fixation, and peak saccade velocity), the study classifies fatigue levels into three categories: low, moderate, and high. Three classifiers are compared: Naïve Bayes, k-Nearest Neighbors, and Random Forest. The results show that Random Forest achieves the highest performance with 87 % accuracy, 86.6 % recall, and 86.9 % F1-score. A strong negative correlation (r = -0.74) is revealed between pupil diameter and fatigue level, indicating pupil constriction as a reliable fatigue marker. Limitations associated with remote webcam-based eye-tracking (lower sampling rate and reduced accuracy compared to lab-grade equipment) are discussed. The findings confirm the feasibility of real-time cognitive fatigue monitoring for adaptive e-learning systems that adjust instructional content based on the learner’s mental state.

The development of rural tourism is urgent to Russia, which makes the related phenomena interesting for the scientific treatment. The objective of the present work is a study of the integration of goat farming and tourism via analysis of the content of a sample of cases. The material is the information extracted from advertisements linked (directly or indirectly) to goat farming and leisure-recreation services. A total of twenty cases was established. The content-analysis procedures include the determination of the principal profiles of entrepreneurial activity, the delineation of tourism elements, and the score-based evaluation of the significance of the latter in the general activities. It is established that the integration of goat farming and tourism appears widely in the country’s space. It is based either on goat farming and production and selling of its products or pure recreation. The main tourism elements are tasting goat milk and related products, master classes, leisure events, photo sessions with goats and excursions. Generally, the studied cases demonstrate a dichotomy. A part of them is related to so-called «primitive» agritourism (leisure-recreation services supplementary to farming), and another part is related to the special development of rural tourism. Such a dichotomy seems to be practically useful, and the integration of goat farming and tourism shows the evident potential. The latter is especially notable due to the perspectives of clustering rural, gastronomic and health tourism. Moreover, the studied cases indicate on the positive effects from the integration of goat farming and tourism for strengthening the traditional values of the Russian society, including those related to family and labor.

The article examines institutional and price distortions of competition in the system of public procurement of the Russian Federation. Based on a comparative analysis of domestic scholarly approaches to the study of competitive processes within the contract system, key factors contributing to the decline in competition are identified. Using data from the Unified Information System and official statistics for the period 2020-2025, it is established that anti-competitive activity has shifted from aggressive price dumping toward “soft collusion”, aimed at maximizing margins while maintaining minimal formal price reductions that preserve the appearance of competitive bidding. A set of measures is proposed, including the introduction of automated “soft collusion” indicators at the procurement planning stage to eliminate structural barriers.

The article outlines the cultural, historical and technological prerequisites, features and factors of structural and substantive changes taking place in modern world tourism and related industries. It is proved that digitalization acts as an unconditional catalyst for changes in the behavior patterns of consumers of tourist services, as evidenced by the results of research conducted by the World Tourism and Travel Council, as well as the World Tourism Organization. The generalization of existing foreign experience in terms of the transformation of consumer behavior convincingly indicates the shift of a number of actions for organizing tourist trips to an online format - starting from the choice of routes and practical travel logistics, the choice of accommodation facilities, ending with the exchange of impressions, assessments, and emotions of tourists.

The Bulgarian National Revival is one of the historical periods in Bulgaria that has been extensively studied by historians, both from the perspective of national liberation movements and in terms of the socio-economic transformations that occurred in various spheres of public life. This paper examines the development of accounting practices in the Bulgarian lands during this period by outlining the distinct stages through which accounting evolved during the Bulgarian National Revival, together with the related terminological changes and the transformation of accounting documents. The study also highlights the relationship between historical events and the development of accounting within this historical context. The article provides a historical overview of key events and presents Turkish lexical borrowings that were used in accounting practice in the Bulgarian lands at that time and formed part of the contemporary culture and social life. These terms are presented in Bulgarian to preserve their authenticity and are explained in Russian to facilitate readers’ understanding of their meaning.

In the context of the active development of the digital retail sector and the involvement of digital retail facilities in retail services, the indicators used in practice to assess the state of retail trade are losing their socio-economic significance. In the article, the author presents a methodological approach to the assessment of retail services, based on the application of the principles and methods of calculating the integral indicator of retail services, taking into account the relationship between the indicators of traditional and digital retail segments.

In today’s context of geopolitical instability and accelerated energy transition, defining strategic development vectors for the oil industry plays a significant role in shaping the industry’s sustainable development. This article proposes a mechanism for mitigating and eliminating risks and threats in the oil industry as a basis for defining strategic development directions for this market segment. Energy foresight served as a link between the risk management process and scenario-based development of the oil industry. Energy foresight is a key element in developing a long-term strategy, ensuring the sustainable development of the oil industry. As part of the energy foresight, a map of oil industry development trends was developed by identifying short-, medium-, and long-term periods and low, medium, and high levels of influence of trends on the oil industry’s development. The study also developed a Roadmap for the Development of the Russian Oilfield Services Industry. A forecast for reducing import dependence in oilfield services is presented, based on three development scenarios: stabilization, modernization, and transformation. The creation of a National Innovation and Technology Center for Oilfield Services Development is proposed, which will ensure the intensive development of oilfield services in the Russian Federation. The presented strategic vectors for the development of the oil industry and oilfield services will contribute to the sustainable development of the fuel and energy complex.

This article is devoted to the identification and assessment of the consequences of the introduction of a new document in the field of accounting regulation. The consequences were identified by identifying uncertainties and subjectivity in the application of the provisions of this document as the causes of distortion of information used in the development and justification of various management decisions. For this purpose, the regulatory acts of the legislation on accounting and related branches of law were studied. As a result, uncertainties have been identified, as well as collisions and conflicts between this document and other regulatory acts in this area.

The article examines the impact of digital transformation on the banking sector through the lens of the introduction of artificial intelligence technologies. The scientific novelty lies in the substantiation of the author’s methodological approach to assessing borrowers’ creditworthiness, based on the integration of explicable AI (XAI) algorithms and analysis of behavioral patterns from open digital sources. Barriers and risks of implementation have been identified, and an algorithm for minimizing regulatory restrictions has been proposed.