The aim of this study is to determine the phase forms of gold in the oxidation zone of the Arkharly deposit and to assess their influence on the efficiency of gold and silver recovery. The main research methods included mineralogical analysis, electron microprobe studies, and technological testing. The study identified and characterized silver halides, established their genetic relationship with gold, and substantiated the role of supergene processes in the redistribution of gold. The Arkharly gold-silver deposit (Kazakhstan) is characterized by a complex composition of ores and a variety of gold occurrence forms, which significantly affect processing efficiency. This paper examines the phase forms of gold in the oxidation zone and their impact on technological recovery indicators. Based on mineralogical, electron microprobe, and technological studies, it was found that gold occurs in various forms: native gold of different fineness, electrum, as well as finely dispersed particles associated with silver minerals. Particular attention is given to silver halides, widely developed in the oxidation zone and forming complex intergrowths with native silver and gold. Textural and structural features of mineral aggregates indicate the redistribution of gold under supergene conditions. The formation of gold-bearing phases is associated with chloride and, presumably, colloidal migration of matter, leading to the formation of zonal structures and inclusions of high-fineness gold. Technological test results showed low efficiency of gravity concentration due to the fine-dispersed nature of gold, and high efficiency of cyanidation and sorption leaching (gold recovery up to 93.5%). A direct relationship between the phase forms of gold and recovery indicators has been established. The obtained results expand the understanding of gold behavior in the oxidation zone and can be used in developing effective technologies for processing complex gold-silver ores.
Project management has primarily developed within high-income countries. However, over the last 15 years, capital-intensive projects have been shifting toward developing nations where conventional frameworks often underperform. Despite this, research on the critical success factors (CSFs) and risks influencing project outcomes in developing countries remains fragmented. This paper conducts a systematic literature review of 41 peer-reviewed articles published in the Scopus database in the last decade, covering 20 developing countries and 8 cross-country studies. Following the PRISMA protocol, this study identifies CSFs and risks in project management (PM) in developing countries through inductive review and deductive coding based on two theoretical lenses, namely, institutional and contingency theory. Furthermore, this paper analyzes the mechanisms and contextual dependency of these factors. The inductive findings identify 22 CSFs and 21 risk factors. In order to report the most critical ones, 8 CSFs and 8 risks are retained based on their frequency of occurrence. The results reveal that they form systematic corresponding pairs, with six of the eight CSFs directly mapping onto six risks. Moreover, institutional theory (IT) explains the dominance of external influences through coercive, normative, and mimetic mechanisms, as well as institutional voids, while the contingency theory (CT) shows variable outcomes across different project characteristics. Overall, the findings provide practical implications for project managers, policy makers and business organizations on managing projects in emerging countries. It is noted that project strategies should be adapted to project scale, with stakeholder-focused approaches in small projects, institutional building in medium projects, and macroeconomic and political risk management in large projects.
This article examines the pressing issue of developing a secure, specialized online platform for distance learning for children with special educational needs. The digitalization of education has opened up new opportunities for inclusion, but mainstream solutions often fail to address the specific needs of this category of students, creating digital, cognitive, and social barriers. The goal of the study is to develop a conceptual model of a secure and adaptive educational environment that comprehensively addresses accessibility, personalization, and cybersecurity. The project took into account the usability of children with various developmental disabilities, including sensory impairments, autism spectrum disorder (ASD), and attention deficit hyperactivity disorder (ADHD), and based on this, the key principles of user interface and user experience (UI/UX) design were formulated. The proposed platform architecture includes an intelligent content and interface adaptation system, personalized learning paths, and secure communication modules with pre-moderation functions. Particular attention is paid to a multi-layered security system that ensures the protection of personal data, the prevention of cyberbullying, and access control. The article is of practical value to educational technology developers, educators, and administrators of educational institutions seeking to create an inclusive digital learning environment.
This study addresses the urgent need to analyze digital threats in everyday discourse by constructing a 1,000text annotated corpus from social media and news platforms covering military and geopolitical events. The purpose of the proposed study is to address the urgent need to analyze digital threats in everyday discourse by creating an annotated corpus of texts with elements of information operations. A multi-layered annotation scheme captures semantic actors and pragmatic features – including impact type, emotional tone, disinformation markers, and intent (e.g., provocation, intimidation). Annotation via Label Studio ensured flexibility, quality control, and context sensitivity, with inter-annotator reliability (Cohen’s Kappa = 0.82) confirming consistency. In pilot experiments, the Onto-IO-BERT model achieved an F1-score of 0.81, outperforming baseline classifiers. Practical utility was validated through analysis of real Telegram messages. The framework is tailored for studying military information operations within Kazakhstan’s Ministry of Internal Affairs and the created corps is a new resource for analyzing military information operations, filling a significant gap in the existing data set. The presented corpus contains texts in Kazakh, Russian and English. The corpus is openly accessible at: https://github.com/baiangali/multi_mil
This work is intended to study methods for pre-processing and analysis of fundus images for the detection of diabetic retinopathy. Diabetic retinopathy (DR) is a common eye disease in patients with diabetes, and its early diagnosis allows you to prevent vision loss. During the study, modern methods for processing and analyzing fundus images were used, including the EfficientNetB0 architecture based on Deep Learning. Image augmentation (rotation, scaling, cropping, contrast enhancement) and normalization methods were introduced for pre-processing. When using the EfficientNetB0 architecture, two approaches were tested: training the base layers and additional adaptation (fine-tuning) by opening the upper layers. The results were evaluated by metrics. The precision for the test set in the first method was 65%, and for the second method 75%. The accuracy of the validation set in the first method was 63%, and in the second method it reached 71%. The recall metric showed 60% for the test set in the first method, and 74% in the second method. In general, the fine-tuning method showed high performance. The use of these methods allows to improve the quality of image processing and classification for effective diagnosis of diabetic retinopathy. The novelty of the study is the analysis of various methods of using and adapting the highly efficient EfficientNetB0 architecture. The results obtained allow to improve the quality of automated systems in DR diagnostics and increase the energy efficiency of the model. The proposed methods have high potential for early detection of eye diseases.
Agriculture is becoming increasingly demanding due to climate change challenges, necessitating continuous monitoring and changes, including soil assessment for precision agricultural requirements. Agricultural soils are heavily utilized by farmers through the application of pesticides and nitrate phosphates to enhance yield. The exacerbation of flood-drought conditions is resulting in soil irregularity, necessitating meticulous soil monitoring at each location. Soil monitoring is prohibitively costly for numerous farmers. To address this issue, the implementation of a compact, energy-efficient, low-cost mobile robotic platform equipped with various sensors for soil monitoring would be prudent. Farmers can remotely manage, analyze surface upper soil strata, and examine topography. The relevance to research activities and active recreation may result in a low cost for a series of behaviors that enhance comprehension of the examined environmental details. The specialized three-wheeled mobility platform is a novel apparatus engineered for autonomous navigation and task execution. The robot’s three-wheel design confers exceptional mobility and stability, enabling effective operation in restricted areas and across various terrains. It is outfitted with sensors and a control system that guarantees accurate navigational control and obstacle evasion. The programming and modification features enable the robot to be tailored for specialized functions, including data collecting, small load transfer, and environmental monitoring. The robot is applicable for educational, scientific, industrial, and domestic uses. Consequently, the three-wheeled mobile robot serves as a versatile and promising platform for the advancement of contemporary robotic systems.
The demand for intrusion detection systems (IDSs) that can promptly identify both known and new types of attacks is on the rise due to the rapid expansion of cyber threats and the consequent increase in network traffic. The utilization of machine learning techniques to autonomously analyze the behavior of network packets and classify them as normal or malicious is a promising way to address this issue. The objective of this investigation is to assess the suitability of a variety of machine learning algorithms for the resolution of network security issues by employing network data analysis as an illustration. This investigation assesses the efficacy of machine learning models in detecting network intrusions using the UNSW-NB15 dataset. This study’s primary objective is to assess the effectiveness of various machine learning models, including Random Forest, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), XGBoost, LightGBM, and Logistic Regression, in network security applications. According to the analysis, all models exhibited high classification accuracy; however, the LightGBM model attained the most remarkable results. This model exhibited the highest values of Accuracy (95.86%), Precision (96.02%), and F1-measure (96.99%), confirming its capacity to effectively manage complex and heterogeneous data. Overall, the study underscores the significance of selecting the most appropriate model based on the security system’s objectives and the specifics of the data.
In this article, we study the expansion of a structure by adding a new predicate that is not definable by any formula in the original language. To consider an externally definable expansion, we define the extension of a model in both essential and non-essential case. Such expansions can lead to significant changes in the properties of the resulting structure. We focus on the case of externally definable expansions, where the new relation is given by the intersection of a formula defined in an elementary extension with the original structure. The concept of a uniformly externally definable expansion was first introduced by Macpherson, Marker, and Steinhorn in the context of expansions by cuts in submodels of o-minimal structures over the real numbers. Subsequently, Baizhanov demonstrated that expanding a model of a weakly o-minimal theory by a family of convex sets preserves both weak o-minimality and uniform external definability. We establish conditions for external expansions under which the key properties of the original structure are preserved.
In the field of speech recognition, end-to-end models are gradually replacing traditional and hybrid approaches. Their main principle is autoregressive decoding, where the output sequence is formed from left to right. However, it has not yet been proven that this method provides the best results in converting speech to text. Moreover, end-toend models rely solely on the previous context, which complicates the processing of unclear or distorted sounds. In this regard, the insertion method was proposed, which does not use autoregressive decoding and generates output data in an arbitrary order. This paper examines a Kazakh speech recognition model trained using the insertion method and Connectionist Temporal Classification (CTC). The experiments conducted showed that this method improves recognition accuracy. Unlike autoregressive models, the Insertion method provides greater flexibility in processing sequences, as it does not require a strict order for generating output data. This reduces decoding delays and makes the model more robust to poorly pronounced words. Furthermore, combining the Insertion method with CTC improves the alignment of audio data and text transcription. This is especially important for agglutinative languages such as Kazakh. According to the experimental results, the recognition accuracy of the proposed model reached 10.2%, making it competitive today.
The article is devoted to the design and configuration of a secure network gateway for cloud applications based on modern VPN protocols OpenVPN and WireGuard. In the context of the rapid development of cloud technologies and the increasing number of cyberattacks, ensuring secure remote access to services has become a key task of information security. The paper discusses relevant threats arising during data transmission in cloud environments and highlights the role of VPN technologies in preventing attacks. The features of OpenVPN and WireGuard are analyzed in detail, including their architecture, cryptographic foundation, ease of configuration, and performance. The study presents a gateway architecture comprising a VPN server, firewall filters, and routing mechanisms that enforce mandatory transmission of all traffic through an encrypted tunnel. Experiments conducted in a virtualized VMware Workstation environment showed that WireGuard provides higher data transfer speeds and lower latency, while OpenVPN demonstrates flexibility and compatibility with corporate systems. The combined use of both protocols improves system resilience and adaptability. The practical significance of the research lies in the possibility of implementing the proposed architecture in corporate and private networks to protect cloud applications, organize secure remote employee access, and enhance the security level of information resources.
This article presents a hybrid machine learning model designed for soil type classification based on the analysis of geophysical characteristics. The proposed model combines two algorithms – RandomForestClassifier and MLPClassifier – integrating the high accuracy of ensemble methods with the ability of neural networks to capture complex nonlinear dependencies between parameters. The input dataset included indicators such as electrical conductivity, density, P-wave propagation velocity, and burial depth. Prior to training, data preprocessing was performed, including outlier removal, standardization, and categorical feature encoding. The hybrid architecture allowed the integration of results from both models with different weights, optimizing classification accuracy. The effectiveness of the proposed approach was compared with alternative algorithms such as XGBoost and Keras using metrics including Accuracy, F1-score, Precision, and Recall. The hybrid model achieved an accuracy of 96.07%, outperforming individual algorithms. Visualization of confusion matrices provided insights into class distribution and model robustness. The results confirm that combining ensemble and neural methods ensures more stable and reliable predictions when working with geophysical data. The developed model can be effectively applied in geotechnical studies, construction, agriculture, and environmental monitoring, enhancing analytical efficiency and reducing the need for costly laboratory testing.
This paper presents a numerical approach for solving Fredholm integral equations of the first kind using the Bubnov–Galerkin method with Alpert wavelet bases. These equations are well-known for being ill-posed, meaning that small changes in input data can lead to large deviations in the solution. Therefore, robust and accurate numerical methods are essential. The proposed method utilizes orthonormal and compactly supported Alpert wavelets, which offer excellent localization properties and yield well-conditioned, sparse system matrices when projecting the integral operator. This enhances numerical stability and reduces computational complexity. A series of computational experiments was carried out using various refinement levels and polynomial degrees. The accuracy of the method was evaluated by comparing approximate solutions to the exact analytical solution. The results demonstrate exceptionally small absolute errors, often approaching machine precision. Additionally, a comparative analysis with power polynomial bases confirms the superiority of the Alpert wavelet approach in terms of convergence and approximation quality. Overall, the method proves to be efficient, stable, and suitable for further extension to more complex integral equations, including multidimensional and noisy-data problems. This confirms the potential of Alpert wavelet-based Galerkin schemes as a reliable tool for the numerical treatment of inverse and ill-posed problems in applied sciences.
In the era of accelerating climate change and growing urban populations, the frequency and severity of natural disasters have increased significantly, posing substantial threats to infrastructure, economic stability, and human lives. Disasters, including the likes of earthquakes, floods, and hurricanes usually are the reasons for serious destruction of buildings, requiring rapid and accurate assessment to aid in emergency response and resource allocation. In light of this, the research aims to deliver a deep learning based building damage assessment model, which is a hybrid architecture consisting of Artificial Intelligence and IOT. In this paper we will examine the use of Internet of Things (IoT) and Artificial Intelligence in disaster management systems in order to improve the automation, transparency, and sustainability in smart intelligence systems. The system should collect and analyze pre-disaster and post-disaster aerial imagery to classify buildings into damage categories, i.e. from no damage to destroyed.. Also, we integrate our model into a wide disaster management system in order to make a visualization of damages on a geospatial interface, that helps the decision-makers to get a quick look at priority areas and streamline the response of disaster. This system’s plan is to assist public authorities, NGOs, and first responders with quick decision making in postdisaster response times.
This scientific article investigates the structure and surface properties of Ti-Au-based thin coatings obtained by the PVD (magnetron sputtering) method in a vacuum environment at a high temperature of 450°C under direct current (DC) and radio frequency (RF) modes. The thin coatings deposited using the NanoPVD system had an average thickness of about 1 µm. Experimental studies showed that the chemical composition, microstructure, and properties of the coatings are closely dependent on the technological parameters of the magnetron sputtering process. The effect of Ag and Cu additions on the microstructure and mechanical properties of the Ti-Au coatings was studied using scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX), and X-ray diffraction (XRD). The surface roughness characteristics of the coatings were analyzed by atomic force microscopy (AFM), and nanoindentation and tribological tests were carried out to comprehensively evaluate the changes in their properties. It was shown that the modification of the Ti6Al4V substrate surface by the PVD method increases its wear resistance and reduces the coefficient of friction. X-ray phase analysis results revealed that the improvement in the tribological properties of the obtained thin coatings is directly related to the formation of the primary Ti3Au phase and the increase in its content.
This study presents results of high-resolution spectroscopic observations of the star MWC 645, a representative of the poorly studied FS CMa-type objects. These stars are characterized by strong emission lines and significant infrared excess caused by circumstellar dust. For the first time, a cool secondary component has been detected in this system. Fundamental parameters were determined for both components: for the hot (B-type) star, an effective temperature of 18,000 ± 2000 K and luminosity log(L/L☉ ) = 3.9 ± 0.4; and for the cool (K-type) star, Teff = 4250 ± 250 K and log(L/L☉ ) = 3.1 ± 0.4. The system is located at a distance of 6.5 ± 0.9 kpc and shows clear signs of active interaction, including ongoing mass transfer indicated by the complex profiles of many emission lines. The results confirm the binary nature of MWC 645 and its classification as an FS CMa-type object. This work highlights the need for further observations to refine the system’s parameters and improve our understanding of the structure and origin of its circumstellar environment.
As cyber threats become more complex, traditional vulnerability detection methods lose their effectiveness. The purpose of this work is to develop and test an approach to identifying vulnerabilities based on the analysis of data from thematic Internet resources: forums, blogs and social networks. These sources contain a large amount of unstructured information, which requires the use of data mining methods. The work uses the integration of modern technologies: the pre-trained SecBERT language model (Security Bidirectional Encoder Representations from Transformers), designed for cybersecurity tasks, and the adaptive neuro-fuzzy inference system DENFIS (Dynamic Evolving Neural-Fuzzy Inference System). The proposed system allows you to filter irrelevant messages, highlight indicators of compromise and potential threats. The use of fuzzy logic makes it possible to efficiently process vague and incomplete information. Experiments confirmed high classification accuracy and stable fuzzy clustering performance (FPC = 0.93; PE = 0.28; XB = 0.042). The system demonstrated the ability to promptly detect signs of cyber threats and has scalability potential for monitoring and attack prediction tasks. The results indicate its potential in increasing the speed of response to cyber threats and strengthening the protection of information systems.
We study a three-weight inequality for a superposition of the Copson, Hardy, and Tandori operators. The goal of this paper is to prove a complete characterization of the boundedness of the operator that is a combination of these three operators in weighted Lebesgue spaces from to . The main focus is on determining necessary and sufficient conditions under which this inequality holds for all non-negative measurable functions on the positive real axis. The notion of a fundamental function of a Borel measure with respect to an increasing function is used substantially. Since the Tandori operator is not a linear operator, we cannot use the duality methods used in earlier works. To solve this problem, we develop a new, simplified discretization method that avoids the complexities of previously known methods. An explicit form of the best constant in the inequality is obtained, demonstrating the accuracy and optimality of the results. By establishing necessary and sufficient conditions for the boundedness of these composite operators, we improve the inequalities previously established in the works of Gogatishvili A., Pick L., Opic B. [1]. The results obtained in the paper extend and complement existing research in the field of weighted inequalities and operator analysis in function spaces and offer potential applications in approximation theory, harmonic analysis and related areas.
This paper presents an automated method for generating the parameters of linear functions used in the diffusion layer of block symmetric encryption algorithms. The focus is on designing linear layers constructed solely from cyclic shift operations and bitwise XORs, which are both efficient and hardware-friendly. Such layers play a critical role in achieving strong diffusion, a fundamental cryptographic requirement. The proposed method evaluates candidate configurations by exhaustively enumerating shift values, calculating their branch number, and assessing their avalanche characteristics. A set of quantitative diffusion metrics is introduced to guide the selection process, including single- and multi-round avalanche effects and activation rates at the byte level. An aggregated quality function is formulated to allow comparative assessment. The developed software tool identified optimal shift parameters for 128-bit blocks processed as four 32-bit words, achieving a branch number of 5 with only 12 XOR operations. The proposed approach contributes to the practical synthesis of lightweight and secure cryptographic primitives suitable for both classical and constrained platforms.
Polymer flooding is one of the key technologies for enhancing oil recovery. Partially Hydrolyzed Polyacrylamide (HPAM) is widely used due to its excellent viscosity-increasing properties. However, the adsorption and retention behavior of HPAM in reservoir porous media presents a dual effect: on one hand, it improves sweep efficiency by increasing flow resistance; on the other hand, it leads to a loss in effective polymer concentration and viscosity, reducing displacement efficiency and increasing costs. Therefore, a systematic understanding and control of HPAM adsorption behavior are crucial for improving the effectiveness of polymer flooding. This work systematically reviews seven main measurement methods for HPAM adsorption quantity, comparing their applicable conditions and limitations. It summarizes the key factors influencing HPAM adsorption and retention behavior from three aspects: polymer properties, rock mineral characteristics, and reservoir environmental conditions. Furthermore, it outlines chemical anti-adsorption methods, represented by competitive adsorption and nanofilm protection, along with their mechanisms. Finally, future research directions are proposed, focusing on building adsorption prediction models, deepening the understanding of adsorption mechanisms under multi-field coupling conditions, and developing novel functional polymers with anti-adsorption capabilities.
Integration of sustainability in construction projects control systems has emerged as an urgent need due to the growing pressure in the environment and the resources toward development of infrastructure projects globally. Whereas traditional Earned Value Management (EVM) has been embraced to track cost and schedule, the performance, it fails to integrate environmental and resource-efficiency indicators that are becoming more significant in affecting the project outcomes. This paper suggests a comprehensive EVM framework, which incorporates three sustainability measures of Carbon Emissions (CE), Energy Consumption (EC) and Material Waste (MW) in the project performance evaluation. The Multiple Linear Regression (MLR) model used, based on nine months of actual operational history of Malir Expressway project in Karachi, Pakistan which is an ongoing large-scale public infrastructure project. The findings indicate that the variables of sustainability play a significant role in determining the Cost Performance Index (CPI) and Schedule Performance Index (SPI). The suggested model enhances the decision-making process because the project stakeholders can monitor the financial and environmental aspects at the same time. Its methodology is practically applicable to the public infrastructure development and will be a base point to the future development of Sustainable Earned Value Management frameworks.
