Bulletin of Abai KazNPU. Series of Physical and Mathematical sciences

FINITE-DIFFERENCE SCHEME FOR INITIAL-BOUNDARY PROBLEM FOR EQUATION OF DIFFUSION WITH VOLTERRA MEMORY

Zh. Abdiramanov — 2025-12-27

An initial–boundary value problem for a nonlinear diffusion equation with a Volterra-type memory term in time is investigated. The model describes diffusion processes in media with hereditary (long-memory) effects and covers, as particular cases, a range of applications in heat and mass transfer. For its numerical solution we construct an implicit two-level difference scheme with a quadrature approximation of the memory integral. A discrete energy identity is derived, which yields energy stability of the scheme under natural constraints on the time step. The nonlinearity is treated by an inner iterative linearization procedure; at each iteration a tridiagonal system of linear algebraic equations arises and is solved efficiently by the sweep (Thomas) method. The convergence of the scheme is rigorously justified and its convergence rate is estimated in the corresponding energy norm. Numerical experiments are presented that confirm the theoretical error bounds and demonstrate the accurate reproduction of long-memory effects.

SOLVING SYSTEMS OF SECOND-ORDER DIFFERENTIAL EQUATIONS FOR SOME MULTIDIMENSIONAL HYPERGEOMETRIC FUNCTIONS

B.K. Abdrakhman — 2025-12-27

In this paper, we study the problem of finding solutions to a system of second-order partial differential equations related to hypergeometric functions of four variables F₁₆, F₁₈, F_19,F₂₀ and F₃₁. Currently, mathematical modeling of physical processes and forecasting their dynamics are becoming increasingly important in science and technology. Most of these processes are described by differential equations or their systems, but only a few of them, those that reflect real physical phenomena, allow analytical solutions expressed in terms of elementary functions. This fact makes it particularly relevant to study more general classes of functions, in particular hypergeometric functions of many variables. Hypergeometric functions are solutions of certain systems of differential equations and belong to the category of special or transcendental functions. Due to their properties, they are widely used in solving multidimensional problems, and the study of their analytical structure is important both for the development of theoretical mathematics and for practical calculations. Within the framework of this study, linearly independent solutions of these systems were constructed for the functions under consideration, as well as their structural features and main characteristics were analyzed. The results obtained form the theoretical basis for further study of hypergeometric functions of a higher number of variables and open up prospects for their effective use in solving applied problems. The work contributes to the development of the theory of special functions and offers new approaches to solving fundamental problems of mathematical physics and applied mathematics.

COMPARATIVE ANALYSIS OF PHYSICS-INFORMED NEURAL NETWORKS AND THE RUNGE-KUTTA METHOD FOR SOLVING ONE-DIMENSIONAL EULER EQUATIONS

S. Alpar — 2025-12-27

This study presents a model describing the physical process associated with gas flow through pipelines using a system of one-dimensional Euler equations. In this work the numerical solution of the system of Euler equations through physically informed neural networks is used, a method that combines modern approaches of deep learning models together with the physical conditions of the problem, which specify the differential equations, initial and boundary conditions. For the validation of the model, the analytical solution of the system of one-dimensional Euler equations will be used and compared with the numerical solution obtained by the deep learning model in the form of graph comparison and metrics calculation. Also, for the relative comparison of this model will be compared with the classical 4-th order Runge-Kutta (RK4) numerical method, where the deep learning method showed higher quality compared to the classical method. The obtained model is of high quality and can be used in gas flow modeling, and the quality of the deep learning model was significantly higher in relative comparison with the classical RK4 method, which emphasizes the effectiveness and novelty of using deep learning methods in comparison with classical methods.

SOLITONS IN THE COUPLED GENERALIZED NONLINEAR SСHRODINGER EQUATIONS

N.S. Serikbayev — 2025-12-27

Nonlinear equations essentially describe physical problems. Solitons, which are localized in space and time and propagate in a nonlinear medium, are the special specific solutions to some of these equations. The soliton remerges after a fully nonlinear interaction, retaining their identities with the same speed and shape. It has remarkable stability properties. Stability plays an important role in soliton physics. The nonlinear Schrodinger equation, one of the soliton-type nonlinear equations, is notable for its significance in the theory of nonlinear waves, specifically in nonlinear optics and plasma physics. The investigation of nonlinear Schrodinger equation generalizations is currently a pressing area of research. This work examines the coupled generalized nonlinear Schrodinger equations. These equations yield classical nonlinear Schrodinger equation in some reductions. As a research method, we use the bilinear Hirota method, which is one of the effective methods for solving nonlinear equations. Dispersion relations are derived and one-soliton and two-soliton solutions are obtained. The figures depict the dynamics of soliton solutions. From this research, the results obtained may be useful for a better understanding of the nonlinear wave phenomena in any varied instance where the coupled model considered is applicable.

APPLICATION OF CORRELATION-REGRESSION METHOD FOR RISK MANAGEMENT IN HIGHER EDUCATION INSTITUTIONS

A. Seriktayeva — 2025-12-27

The purpose of the work is to assess current risks in universities using mathematical methods. The decline in the number of applicants to regional universities of the country is one of the main risks in the strategic development program. The conditions that could affect this risk were identified, and a regression analysis was performed based on the data obtained. For completeness of the results obtained, the correlation and regression method was used. A unified system of indicators has been created through the signs and features of economic, social, personal and information factors of accessibility of higher education. As a result of the study, a regression equation was created and the amount of risk was calculated, the significance of the constructed models was assessed, and the interpretation of the results obtained in the modeling process was given. The theoretical and methodological significance of the study lies in improving the methodology by substantiating the possibility of using correlation and regression analysis and economic and mathematical modeling.

NONLINEAR EFFECTS IN WAVE PROCESSES OF POROUS MEDIA

A. Yussupova — 2025-12-27

Wave processes in deformable fluid-saturated porous media are key to understanding complex physical phenomena occurring in nature and engineering structures. Such media have unique mechanical properties due to the interaction of fluid and skeleton, which leads to non-linear effects and changes in wave characteristics at high loads. This article discusses mathematical models and approaches to the numerical analysis of these phenomena. Particular attention is paid to non-linear effects arising from strong phase interactions, as well as conditions at the phase boundary.

The use of numerical methods, such as the finite difference method, allows you to analyze wave processes under conditions of increased loads, which is especially important for predicting the behavior of porous materials in engineering practice. Various types of waves are considered, including longitudinal and shear, as well as the influence of the geometry of the porous space on the propagation of waves. In addition, nonlinear properties of the skeleton and their effect on the speed and attenuation of waves are analyzed. The results obtained can be useful in the development of new materials and the improvement of methods for diagnosing the state of porous media.

THE IMPACT OF THE GEOGEBRA PROGRAM ON IMPROVING THE LEARNING QUALITY OF 6TH GRADE STUDENTS IN TEACHING LINEAR INEQUALITIES

A. Aituar — 2025-12-29

In this paper was considered the effect of the computer interactive program Geogebra on learners academic performance in teaching the system of linear algebraic inequalities. The research involved 30 learners from Mahatma Gandhi Specialized Lyceum no 92 in Almaty. Lessons in the experimental group were conducted using the Geogebra program, and the control group received on the traditional method of teaching. This research was conducted among 6^th-class learners. During the research, learners from both groups were assessed pre and post tests, consequently learners in the experimental group showed significant improvement in academic performance compared to the control group (with an effect size of d = 0.819). In the end of the research, a survey was conducted among learners about the impact of the Geogebra program on learning outcomes. As a result of the research, the effect of the program Geogebra on learners academic performance in teaching the topic of "System of linear algebraic inequalities" was determined

STEM ЖАҒДАЙЫНДА БОЛАШАҚ МАТЕМАТИКА МҰҒАЛІМДЕРІН ӘДІСТЕМЕЛІК ДАЙЫНДАУДА ОҚЫТУҒА АРНАЛҒАН ОЙЫН ВИКТОРИНАЛАРЫ

N.B. Ospanova — 2025-12-27

STEM технологияларының жоғары дидактикалық әлеуетіне қарамастан, оларды педагогикалық жоғары оқу орындарында болашақ математика мұғалімдерін даярлау үдерісіне енгізу белгілі бір қиындықтарды тудырады. Негізгі мәселе оқытудың пәндік форматының ерекшелігінен туындайды: ол тиісті әдіснамалардың болмауына байланысты бірнеше пәнді бір уақытта меңгеруге немесе пәнаралық жобаларды жүзеге асыруға мүмкіндік бермейді.

Бұл жағдайда жағдайда перспективалы шешім ретінде мұғалімдерді байланысты пәндерден алған білімдерін оқу процесіне енгізуге ынталандыратын ойын технологияларын пайдалану ұсынылады.

Мақалада болашақ математика мұғалімдерінің әдістемелік даярлығын қамтамасыз етуде оқу сұрақтық викториналарының тиімділігі негізделеді. Мұндай викториналар оқушылардың танымдық белсенділігін арттырып қана қоймай, сонымен қатар STEM құзыреттілігін дамытуға ықпал етеді.

Викториналардың маңызды артықшылығы — білім алушылардың кіріктірілген білімге деген қызығушылығын ояту қабілеті. Бұл әртүрлі пәндік салалардан алынған сұрақтар мен тапсырмаларды қамту арқылы жүзеге асады.

Оқытушылық және диагностикалық материалды жүйелеу мақсатында Барбара Минтоның пирамида әдісін қолдану ұсынылады. Аталған әдіс дайындықтың әртүрлі бағыттарын қамтитын көпсалалы сұрақтар мен тапсырмалар ағашын құруға мүмкіндік береді. Викториналарды танымал бұлтты сервистерде де, сондай-ақ мамандандырылған конструкторлардың көмегімен де әзірлеуге болады.

SYSTEMATIC ANALYSIS OF THE POTENTIAL USE OF GAMIFICATION ELEMENTS IN THE TRAINING OF FUTURE COMPUTER SCIENCE TEACHERS

A. Aldabergenova — 2025-12-27

This study presents a systematic review of the application of gamification in higher education between 2020 and 2025. The methodological foundation is based on the PRISMA protocol, which ensured a multi-stage search, screening, and structured selection of publications. An analysis of Web of Science, Scopus, Google Scholar, and ResearchGate databases identified 3451 publications, of which 42 full-text studies were included in the final sample after removing duplicates and applying inclusion criteria.

The data analysis employed thematic classification, content analysis, and visualization using Microsoft Excel, VOSviewer, and Datawrapper. The findings indicate that the most frequently used gamification elements in university teaching include points, badges, levels, and leaderboards, which demonstrate consistently positive effects on students’ motivation, engagement, and academic activity. Less common mechanics related to progress, rewards, and competition are used selectively and show more variable outcomes.

Special attention in this study is given to the potential use of gamification in the training of future computer science teachers. The identified elements possess various pedagogical benefits, including the development of motivation, autonomy, active learning behaviors, and problem-solving skills, making them particularly relevant for the methodological preparation of future educators. The practical significance of the study lies in the fact that its results can be used in the development of gamified instructional materials and digital courses in computer science teaching methodology.

THE POSSIBILITIES OF USING AUGMENTED REALITY TECHNOLOGY IN TEACHING ROBOTICS TO SCHOOLCHILDREN

Ye. Zhabayev — 2025-12-27

The article is devoted to exploring the potential of using game elements in teaching robotics within the context of modern STEM education. The use of gamification in the educational process enhances students’ motivation, engagement, and the effectiveness of learning outcomes. An analysis of scientific research has shown that the integration of game mechanics—such as point systems, levels, virtual rewards, and narrative scenarios—has a positive impact on knowledge acquisition, the development of engineering thinking, and the formation of teamwork skills. Special attention is given to the Kazakhstani experience of integrating gamification, particularly through the implementation of the mobile application RoboBil, designed for teaching robotics to middle school students. The application combines theoretical materials, practical tasks, quizzes, and AR-based reward models, enabling personalized learning paths and strengthening the gamification effect. The research results demonstrate that the use of the RoboBil mobile application increases students’ interest in robotics, improves their academic performance, and develops their problem-solving skills. Thus, gamification combined with mobile technologies opens new perspectives for the development of educational robotics and can serve as a model for introducing innovative practices in schools.

DEVELOPING A METHODOLOGY FOR TEACHING 3D MODELING IN SCHOOLS USING ARTIFICIAL INTELLIGENCE RESOURCES AND THE BLENDER PROGRAM

S.A. Zhamalova — 2025-12-27

The article explores the development of a teaching methodology that integrates artificial intelligence technologies and the capabilities of the Blender program to foster three-dimensional modeling skills within the secondary education system. The aim of the study is to organize 3D modeling lessons based on innovative approaches that promote students’ spatial thinking, creativity, and engineering abilities. The article analyzes the theoretical foundations of 3D modeling, its significance in education, and the potential for personalizing and adapting the learning process through the use of artificial intelligence tools. A pedagogical experiment was conducted as part of the research, and its results practically confirmed the effectiveness of the proposed methodology. The findings are aimed at improving digital educational content for school students and integrating modern technologies into the learning process.

METHODOLOGY FOR TEACHING FUTURE COMPUTER TEACHERS OPERATING SYSTEMS USING A VIRTUAL ENVIRONMENT

G.D. Koshanova — 2025-12-27

At present, one of the key challenges in the higher education system is the preparation of competitive and professionally competent informatics teachers. This article provides a theoretical and practical justification for the importance of using virtual machines in teaching operating systems to future informatics teachers. The aim of the study is to develop an effective methodology for teaching operating systems through the use of virtual machines. The article examines the capabilities of virtualization tools such as VirtualBox and describes ways to apply them effectively in the educational process. Practical tasks performed in a virtual environment enable students to safely and flexibly develop skills in operating system management. In addition, the proposed methodology contributes to the optimal use of hardware resources and increases the efficiency of working with complex software. As a result of the study, practical tasks for future informatics teachers were developed and integrated into the educational process. In conclusion, virtualization technologies are proposed as an effective tool for developing students’ practical competencies.

DEVELOPMENT AND IMPLEMENTATION OF IT-RELATED MICRO-CREDENTIALS FOR STUDENTS OF PEDAGOGICAL UNIVERSITIES

N.T. Oshanova — 2025-12-27

In the modern education system, micro-credentials are considered an effective tool for developing students’ flexible skills and adapting them to the requirements of the labor market. However, in pedagogical universities, the experience of developing and implementing IT-oriented micro-credentials remains insufficiently systematized. Most existing studies focus primarily on technical specializations, which leaves the specifics of developing digital competences among students in teacher education largely unexplored. Moreover, based on the Atlas of New Professions, there is a lack of practical recommendations for revising educational programs and adapting international practices. In this regard, the introduction of IT micro-credentials for students of pedagogical universities is a relevant step towards improving the quality of education, expanding access to digital resources, and preparing specialists who are in demand in the modern labor market.

Research aim: to identify the scientific and methodological foundations for the development and implementation of IT micro-credentials for students of pedagogical universities, to assess their alignment with labor market requirements, and to determine pathways for integration into the educational process.

Methodology: the study conducted an expert analysis of IT-related areas outlined in the Atlas of New Professions. In addition, pilot modules incorporated into curricula were examined, and a survey was carried out among students and faculty members. Based on the collected data, a structural model of micro-credentials was developed and implementation mechanisms aligned with labor market needs were proposed.

Results: the study proposed a modular structure of IT micro-credentials adapted for pedagogical universities. This structure fosters the development of students’ digital competences, the formation of labor market-oriented learning experiences, and the integration of international best practices. The suggested solutions contribute to the modernization of educational programs, acceleration of digital transformation, and enhancement of the competitiveness of future educators.

TRAINING COMPUTER SCIENCE TEACHERS TO WORK WITH GIFTED STUDENTS: MODERN METHODS AND TECHNOLOGIES

G.A. Madyarova — 2025-12-27

In today’s education system, under the conditions of digital transformation, identifying and supporting giftedness remains a pressing issue. The study, based on global and domestic experiences, showed that early identification of an individual's creative and intellectual potential contributes to increased education quality and higher adaptability to a competitive information environment. The development of human capital among gifted students and the formation of an innovative society requires enhancing their personal abilities and creative potential, making it one of the key factors.

This article analyzes modern methods and technologies for identifying giftedness in the education system. The works of scholars in this field were reviewed, and content analysis and modeling methods were applied. As a result, three main directions for identifying and developing giftedness were identified: diagnostics, support strategies, and digital competence.

As a result of the research, the Smart-Talent Core («Intellectual Gift Core») model was developed for integrating the processes of identifying, developing, and supporting giftedness in the digital environment. This model is designed to identify and develop students' potential, as well as improve the professional competence of educators in educational institutions (preschools, schools, colleges, and universities).

The research findings contribute to the introduction of innovative approaches to working with gifted children, the digital transformation of the educational process, and the creation of an intellectual educational environment.

EVOLUTION OF USING VIRTUAL AND AUGMENTED REALITY TECHNOLOGIES IN TRAINING INFORMATICS TEACHERS

A. Samat — 2025-12-27

The article analyzes the development trends and pedagogical potential of using virtual reality (Virtual Reality) and augmented reality (Augmented Reality) technologies in the professional training of informatics teachers in recent years. The purpose of the study is to identify the evolution of applying virtual and augmented reality technologies in the process of training future informatics teachers and to determine their role in developing professional competence. The study employed content analysis and system-based methods. Over the past decade, the scope of immersive technologies in education has significantly expanded, and their integration into pedagogical training is considered an effective approach to enhancing the practical and methodological skills of pre-service teachers. Virtual and augmented reality technologies contribute to the creation of a modern learning environment that promotes students’ project-based activity, visual thinking, and digital literacy. As a result of the study, an evolutionary model for using these technologies in informatics teacher training was proposed, consisting of four stages: adaptation, application, integration, and improvement. The practical significance of the model lies in improving the efficiency of digitalizing pedagogical education and organizing immersive learning environments in higher education institutions.

THE NECESSITY OF DEVELOPING PEDAGOGICAL RESEARCH SKILLS IN FUTURE COMPUTER SCIENCE TEACHERS

T.T. Toishybek — 2025-12-27

The purpose of this study is to identify the need to develop pedagogical research skills among future computer science teachers. The objectives of the study include the analysis of modern methods and approaches to the development of these skills, as well as assessment of their impact on the quality of the educational process. The research methodology is based on the use of qualitative and quantitative methods, including surveys, interviews and analysis of teaching practice. The results of the study show that the development of research skills among future computer science teachers helps improve their professional competence, improve the quality of teaching and introduce innovative teaching methods. This study is significant for science, as it highlights the need to integrate research methods into the teacher training process, which in turn contributes to more effective and high-quality education in the field of information technology

APPLICATION OF MACHINE LEARNING AND NLP METHODS FOR ENHANCING PROFESSIONAL APTITUDE

N.B. Abutalipova — 2025-12-27

The article discusses the development and optimization of an artificial intelligence (AI)-based information system aimed at shaping learners' career aptitude. The relevance of using AI is justified in the context of dynamic changes in the labor market and the limitations of traditional career guidance methods. The proposed system architecture includes data collection models via an intelligent chatbot, market data analysis using NLP and clustering techniques, as well as methods for matching learners' profiles with professional requirements. Key AI technologies are described in detail, including natural language processing models (BERT, GPT, Llama), adaptive dialogue algorithms (decision trees, reinforcement learning), and matching methods (cosine similarity, Siamese networks). Special attention is given to the recommendation mechanism that integrates both compatibility and market demand criteria.

APPLICATION OF MACHINE LEARNING METHODS TO ANALYZE MALICIOUS NETWORK TRAFFIC

G. Aksholak — 2025-12-27

The increasing deployment of Internet of Things (IoT) devices has made networks more vulnerable to malicious activities, necessitating effective traffic classification methods. This study investigates machine learning-based approaches to classify network traffic using the CTU-IoT-Malware-Capture-3-1conn.log.labeled dataset, which includes over 150,000 records labeled as benign or malicious. Key numeric features such as packet counts, data volumes (bytes), protocol types, and source IP frequencies were selected to enhance the models' predictive capabilities. Preprocessing involved normalization with StandardScaler to ensure equal contribution of features to model predictions. Four machine learning algorithms were evaluated: Logistic Regression, Random Forest, Support Vector Machines (SVM), and Gradient Boosting. The experiments utilized a 5-fold cross-validation framework to ensure robustness and reliability. Gradient Boosting outperformed other models with a mean accuracy of 99.13%, followed by Random Forest at 99.11%. To address class imbalance, SMOTE was applied, significantly improving the recall of minority classes. This study demonstrates the potential of machine learning for improving IoT network security by identifying anomalous behaviors in real-world attack scenarios. The results underline the importance of preprocessing, feature selection, and evaluation in achieving high detection accuracy.

MATHEMATICAL MODEL OF A PRIVATE UNIVERSITY'S VDI CLOUD IN COMPLIANCE WITH INFORMATION SECURITY REQUIREMENTS

G.B. Yelubay — 2025-12-27

The article offers an extended mathematical model for estimating the optimal number of computing nodes in a private university cloud using virtual workstations (VDI), and taking into account the information security requirements of a cloud-based learning environment (OSS) of universities. Unlike previously known approaches, the developed model takes into account not only the resource load (CPU, RAM), but also the requirements for the security level of virtual machines and servers. In the proposed model, quantitative parameters of the security of the OSS are introduced, and the task of VM deployment is reduced to a multi-criteria optimization balancing between resource efficiency and compliance with the security requirements of the OSS. A simulation simulation of the cloud operation with typical parameters of the university IT infrastructure has been carried out, confirming the practical significance of the proposed approach. The model is suitable for designing secure cloud platforms in the educational environment of universities of the Republic of Kazakhstan.

THE APPLICATION OF METAHEURISTIC METHODS FOR OPTIMIZATION OF DISTRIBUTED IT SYSTEMS

N.S. Yesmukhamedov — 2025-12-27

This paper examines task distribution optimization methods in IT systems using metaheuristic algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO). The study explores their applicability to solve complex problems in modern IT environments, where traditional methods fail to efficiently process large data volumes or provide the required accuracy. The research focuses on IT systems that need optimal allocation of computational and human resources to enhance performance and reduce costs. Mathematical models for dynamic optimization are developed, considering parameters such as cost, time, and quality of task execution. A comparative analysis of the three methods (GA, PSO, ACO) showed that each has its strengths and weaknesses in the context of optimization tasks: GA is most effective in terms of time but with higher costs, while PSO and ACO deliver better results in quality with lower time and memory costs. The practical value of the research lies in the potential application of the proposed methods for automating resource management processes in IT, significantly improving operational efficiency and reducing costs. The scientific value of the work is in expanding theoretical approaches to using metaheuristic methods to solve optimization problems in IT services and project management.

DEVELOPMENT OF A KAZAKH SIGN LANGUAGE RECOGNITION SYSTEM USING COMPUTER VISION AND DEEP LEARNING METHODS

M.S. Zhassuzak — 2025-12-27

This paper discusses the development of a Kazakh Sign Language recognition system and the enhancement of its effectiveness. An urgent task is the creation of a system capable of recognizing Kazakh Sign Language in real-time to facilitate communication among individuals with hearing and speech impairments. The proposed method employs a YOLOv5 convolutional neural network and the Mediapipe library, ensuring high-precision real-time gesture recognition. For the analysis and semantic processing of the recognized signs, a Long Short-Term Memory (LSTM) network is utilized. The developed system allows for the real-time analysis of users' hand movements and the generation of meaningful sentences from the recognized gestures. Furthermore, a web application based on the Django framework was developed to conveniently present the results to users. Experimental results demonstrated that the proposed system provides high accuracy and reliability in recognizing Kazakh Sign Language in real-time.

ANALYSIS OF SOCIAL ENGINEERING THREATS IN SOCIAL NETWORKS AND METHODS TO COMBAT THEM.

Zh.K. Ibrayeva — 2025-12-27

Observing the growing popularity of social networks as a medium of modern communication, which provides users with extensive opportunities for information exchange, there is an unseen rise in the threats of spreading malicious information associated with the use of social engineering methods. This study focuses on the phenomenon of social engineering aimed at manipulating social network users to extract confidential data, damage reputations, spread disinformation, or cause other harm. The relevant key mechanisms of social engineering, the reasons for users' vulnerability to such attacks, and the potential consequences for information security are considered. The work also offers recommendations for countering these threats, including enhancing digital literacy, developing protective technologies, and improving regulatory mechanisms. This research is relevant to specialists in the field of information security, social sciences, and anyone interested in protecting the digital environment from manipulation and malicious information.

RECENT ADVANCEMENTS IN SKELETON-BASED SIGN LANGUAGE RECOGNITION

T. Sembayev — 2025-12-27

Sign language recognition is vital for bridging communication between deaf communities and the hearing world, yet automatically decoding both static and dynamic gestures remains challenging. Over the past decade, skeleton-based approaches—where only joint coordinates (2D or 3D) are extracted from video—have emerged to reduce computational demands and preserve signer privacy. In this review, we examine nineteen key studies from 2015 to 2024, selected by PRISMA 2020 guidelines via a Scopus search, that employ deep models (ST‐GCN, GCN+RNN hybrids, Transformers) trained on skeletal data. We find that while isolated‐word tasks achieve Top‐1 accuracies of 98–99 % (e.g., on AUTSL), performance for continuous sign streams falls to 80–90 % (word error rates of 10–20 %). Multi‐stream architectures—fusing joint positions, bone vectors, and motion cues with attention mechanisms or Transformer layers—offer the best gains, boosting dynamic sign recognition by 3–5 %. Key limitations include the lack of large, richly annotated skeletal corpora, under‐representation of non‐manual signals (facial expressions, torso motion), and limited model portability across different sign languages. We conclude that skeleton‐based methods hold strong potential for privacy‐preserving, noise‐robust gesture recognition; yet realizing real‐time continuous translation will hinge on expanding dataset diversity, tightly integrating non‐manual articulators, and developing hybrid architectures capable of handling the full complexity of live sign language.

ALGORITHMS FOR DEVELOPING EDUCATIONAL PROGRAMS FOR USE IN INFORMATION SYSTEMS

A. Serikov — 2025-12-27

This paper presents an algorithmic framework for educational program development in modern information systems, bridging traditional curriculum design with dynamic labor market demands. It introduces the Design Developer of Educational Programs (DDEduP), a web-based platform integrating a fine-tuned Llama 3 model with retrieval-augmented generation (RAG). The system draws real-time data from Kazakhstani job platforms (Enbek.kz, HeadHunter), the Atlas of New Professions, and professional standards, using semantic embeddings, cosine similarity, and hybrid retrieval to generate goals, extract competencies, and recommend courses. Evaluated across Computer Science, Biology, and History, it achieved F1 scores from 0.24 to 1.00, excelling in goal formulation and technical fields but facing challenges in diverse learning outcomes and interdisciplinary adaptation. The study advocates human-in-the-loop integration and discipline-specific refinements for effective AI-driven curriculum design in educational systems.

DEEP LEARNING-BASED CYBERBULLYING DETECTION IN KAZAKH: A HYBRID APPROACH FOR IMPROVED TEXT CLASSIFICATION

D. Sultan — 2025-12-27

This study presents a deep learning-based approach for detecting cyberbullying in the Kazakh language, addressing key challenges associated with low-resource languages. The research highlights the increasing prevalence of cyberbullying in Kazakhstan, the limitations of traditional machine learning models, and the need for advanced text classification techniques. A novel hybrid deep learning model integrating Convolutional Neural Networks (CNNs), Bidirectional Long Short-Term Memory (BiLSTM), and Transformer-based architectures is proposed to enhance detection accuracy. The study details the dataset collection process, data augmentation techniques, and model evaluation using key performance metrics such as accuracy, precision, recall, and F1-score. Experimental results demonstrate that the proposed model significantly outperforms conventional machine learning algorithms and previously published methods. The findings offer practical implications for automated content moderation on social media platforms and contribute to the advancement of natural language processing (NLP) tools for the Kazakh language.

METHODS OF STRESS ASSESSMENT AND PREDICTION BASED ON PHYSIOLOGICAL PARAMETERS

G A. Tyulepberdinova — 2025-12-27

Stress is one of the most pressing problems of modern society, which directly affects a person's mental and physical health. Prolonged or uncontrolled stress is one of the main factors of cardiovascular diseases, weakened immune system and decreased productivity. In this regard, timely detection and prediction of stress is becoming an important scientific and practical task. This article discusses the problem of predicting stress based on physiological indicators. These studies were recorded using a specially designed instrument that allows you to collect parameters of heart rate (HR), respiratory rate (RR) and skin conduction (SC). Based on the received signals, an integral Stress Index was calculated, and according to the interquartile percentiles, the stress level was divided into four categories: quiet, low, normal, high. Using this data, machine learning algorithms such as KNN, Decision Tree, Random Forest and Gradient Boosting were used. Experimental results showed that the Random Forest model, which achieved the highest result (hold-out accuracy 0.900, average accuracy of 5-layer cross-validation 0.805), proved to be a stable and reliable method for predicting stress. The results obtained make it possible to carry out real-time monitoring on wearable devices and create systems for early detection and prevention of the user's psychophysiological condition.

EVALUATION OF THE EFFECTIVENESS OF USING THE PORTABLE SATELLITE COMMUNICATION COMPLEX IN HARD-TO-REACH AREAS

K.S. Chezhimbayeva — 2025-12-27

In the modern conditions of providing communications to remote and hard-to-reach areas, satellite systems that provide reliable and high-quality communications regardless of the terrain and infrastructure limitations are of particular relevance. The purpose of the study is to assess the effectiveness of using the PSK-400 satellite complex to organize communications in hard-to-reach areas. The key technical characteristics of the PSK-400 are analyzed, including throughput, coverage area, resistance to interference and signal delay. Scenarios for its use in various conditions are considered: providing communications in mountainous areas, arctic territories, remote industrial facilities and emergency zones. The effectiveness is assessed based on a comparative analysis with alternative solutions, as well as the results of practical testing. The advantages and possible limitations of using the PSC-400 are identified, including economic and operational aspects. The study is aimed at identifying the advantages, limitations and prospects for using this complex in various conditions. The obtained results confirm the feasibility of using the PSK-400 satellite complex for organizing communications in hard-to-reach areas, which can contribute to the development of digital infrastructure and increase the level of information accessibility in remote regions.