Scientific Journal of Astana IT University
Not a member yet
250 research outputs found
Sort by
MODELING THE EFFECTIVENESS OF FPV DRONE OPERATOR TRAINING USING SIMULATORS AND ONLINE PLATFORMS
Get PDFThis article examines the key conditions and factors influencing the training efficiency of FPV drone operators through simulators and online platforms in Kazakhstan. The study aims to address the lack of standardized methodologies and national frameworks for UAV operator training by identifying socio-economic, technical, and pedagogical determinants that shape learning outcomes. Using a mixed-method approach combining literature analysis, comparative assessment of international practices (USA, China, the UK, and Australia), and mathematical modeling, the research formalizes the relationship between simulator-based learning, real flight practice, and external factors. The proposed integrated model E(t) quantifies training efficiency as a dynamic function of simulation-based skill acquisition, reinforcement through practical flights, and the impact of organizational and infrastructural conditions. Results demonstrate that hybrid training pathways – combining intensive simulator preparation with supervised real flights – significantly enhance skill retention and operational safety while reducing costs and training time. Comparative analysis of global ecosystems reveals that advanced training systems increasingly integrate virtual and augmented reality (VR/AR) and artificial intelligence (AI) for adaptive learning and error analytics, whereas Kazakhstan faces challenges of uneven infrastructure development and limited access to standardized resources. The findings underscore the need for national adaptation of international best practices, the creation of domestic simulation centers, and the development of unified educational standards for FPV operator certification. The proposed model and recommendations can serve as a foundation for policy development, simulator design, and academic curricula, contributing to the formation of a skilled workforce and the sustainable growth of the national drone industry
ENSEMBLE MACHINE LEARNING FOR GLOBAL HYDROLOGICAL PREDICTION
No full textAccurate global hydrological prediction is vital for sustainable water management but is often hindered by data complexity and fragmentation. This study introduces an advanced machine learning framework to predict long-term average discharge using widely available global hydrological station metadata, aiming to develop a highly accurate and generalizable model for large-scale water resource assessment. The methodology utilized the Global Runoff Data Centre (GRDC) dataset, applying extensive feature engineering to station characteristics and a logarithmic transformation to the discharge variable. A diverse set of algorithms was trained, including a custom deep neural network with specialized architecture and several gradient boosting machines. These individual models were then integrated into a final Meta Ensemble model through an optimized weighting strategy to maximize predictive performance. The framework was rigorously validated on an independent test set. The Meta Ensemble model demonstrated superior predictive power, achieving a Coefficient of Determination (R²) of 0.954. This performance significantly surpassed that of both baseline methods and the individual advanced models. Analysis of the results confirmed that the model learned hydrologically meaningful relationships, identifying catchment area and geographical location as the most influential predictors. The findings confirm that a data-driven ensemble framework can accurately predict key hydrological characteristics using only station metadata. This approach offers a powerful and scalable alternative to traditional modeling, holding significant potential for water resource assessment in data-scarce regions and serving as a robust foundation for future intelligent monitoring systems
ENHANCED INFORMATION SECURITY FOR VOTING SYSTEM IN EMERGENCIES USING PAILLIER’S CRYPTOSYSTEM
Get PDFThis study explores the application of Paillier’s Partial Homomorphic Encryption (PHE) in the context of secure digital voting systems, particularly in emergency situations such as pandemics, natural disasters, or martial law. The proposed system is implemented using Python with the Django framework and the pycryptodome library to ensure a secure and scalable environment. A key feature of Paillier’s cryptosystem is its ability to perform computations directly on encrypted data, which preserves voter confidentiality and guarantees data integrity without requiring decryption. A simulated voting scenario involving 10 voters and 3 candidates was conducted to evaluate the system. Encrypted votes were processed using homomorphic operations, allowing for the secure aggregation of votes. The results demonstrated that the system accurately computed vote totals—35 votes for Candidate A, 50 for Candidate B, and 100 for Candidate C—without compromising security. The system proved efficient and reliable for small-scale implementations. However, the study identifies significant challenges when scaling the system to national-level elections. The cryptographic operations required by Paillier’s scheme are computationally intensive and could hinder performance when processing millions of encrypted votes in real-time. Therefore, while the system shows high potential for secure e-voting, further research is required to optimize performance. The authors propose future work in two directions: optimizing the underlying cryptographic operations and integrating blockchain technologies to enhance transparency and auditability. Overall, the results suggest that Paillier’s PHE provides a robust framework for emergency e-voting systems and offers a substantial improvement over traditional voting methods in terms of both security and privacy
ARTIFICIAL INTELLIGENCE-ENHANCED MOBILE DIAGNOSTICS USING DECISION TREES FOR EARLY DETECTION OF RESPIRATORY DISEASES
Get PDFThis article is devoted to the identification of early diagnosis of respiratory lung diseases, such as chronic obstructive pulmonary disease and pneumonia, to reduce mortality and prevent complications. One of the most effective methods of structuring data is the Decision Tree model. The research focuses on the development and evaluation of a decision tree model, which is used to obtain data in the form of questionnaires, text files from patients, where they describe in detail the entire process of the disease, describing their symptoms and general condition at different time periods. There are a few criteria that patients must answer for a more accurate diagnosis. The developed methodology will allow processing relevant data with various symptoms to obtain reliable identification of the signs of the disease, as well as the stages of its progression; this can be done without the use of complex and high-tech devices that make diagnosis very accessible and feasible in the shortest possible time, if resources and time are limited. The article describes the model, carefully collected, and processed the necessary data, and then the results will be described in detail, covering many indicators such as accuracy, responsiveness, F1 score and ROC-AUC. The results of this analysis strongly suggest that this model is effective enough to provide a high level of accuracy combined with extensive capabilities, which determines its practical importance for use in real conditions. It is noted that the decision tree model can significantly improve the quality of diagnostics, since it is possible to structure a large amount of data and thus collect many years of human experience
ANALYSIS OF TECHNICAL FEATURES OF DATA ENCRYPTION IMPLEMENTATION ON SD CARDS IN THE ANDROID SYSTEM
Get PDFThis article provides a detailed analysis of data encryption mechanisms for removable storage devices in the Android operating system. Two main information protection technologies are examined: file-based encryption when using an SD card as portable storage and full-disk encryption when using a memory card as an extension of the device's internal storage (Adoptable Storage). The technical implementation features of each method are investigated, including the encryption algorithms used, the structure of encrypted data, and key storage mechanisms. The research was conducted using Sony Xperia XZ and Xiaomi Redmi 5 Plus devices, employing tools for working with file systems and encryption based on Linux and Android. The analysis has established that full-disk encryption is utilized the dm-crypt kernel module in plain mode with AES-256-CBC-ESSIV:SHA256 cipher. The partition encryption key is stored in the device's internal memory. File-based encryption employs the eCryptFS kernel module. The file structure includes information about the original file size, format marker, flags, number of extents, their size, and the encryption key. Comparative analysis has shown that Adoptable Storage mode provides more comprehensive data protection through full-disk encryption, while Portable Storage mode with file-based encryption offers greater flexibility in use but may be less secure due to the possibility of analyzing the file system structure and file metadata. Research has revealed the implementation of encryption mechanisms depends on the device manufacturer and Android operating system version. The research findings have practical significance for understanding the level of data protection using different modes of removable storage operation in the Android system and are useful for both developers and information security specialists, as well as ordinary users
FEATURE SELECTION METHODS FOR LSTM-BASED RIVER WATER LEVEL AND DISCHARGE FORECASTING
Get PDFAccurate forecasting of river discharge and water levels is essential for effective water resource management, flood mitigation, and public safety. This study compares correlation-based and PCA-based feature selection methods for LSTM forecasting models in the study area at Uba River basin, within Shemonaiha city in the East Kazakhstan region. The dataset spans from 1995 to 2021, with 1995 to 2019 used for training and validation and 2020 to 2021 for testing. Both feature selection methods reduced the original predictor set to 13 features while generally maintaining predictive accuracy. An ensemble of 10 LSTM models was trained using 60-day input sequences to forecast discharge and water levels over a 10-day horizon, reducing variance from random initialization and stabilizing predictions. Performance was evaluated using the Nash-Sutcliffe Efficiency. Results showed that correlation-based selection performed comparably to the full-feature baseline in 2020 test set, suggesting that removing highly correlated predictors did not decrease short-term forecasts capacity of the model. The model with PCA-based selected features, while slightly lagging at longer lead times in 2020, exhibited advantages in most lead times with 2021 forecasts. However, overall predictive performance declined in 2021 compared to 2020, indicating that the hydrological conditions deviate more from the historical training record, and suggesting the need for model updates with relevant historical training data. Both feature selection methods successfully reduced dimensionality, while preserving performance capacity, though neither was universally superior across all forecast lead times. These results emphasize the value of systematic feature selection in hydrological modeling and highlight the importance of model adaptability to evolving environmental conditions
DEVELOPMENT AND IMPLEMENTATION OF AN AUTOMATED WEB-BASED KPI MANAGEMENT AND DASHBOARD SYSTEM AT ASTANA IT UNIVERSITY
Get PDFEvaluating Key Performance Indicators (KPIs) of faculty and staff is critical to ensuring accountability and promoting institutional effectiveness in higher education. However, the management of these processes often relies on manual, error-prone systems, creating significant administrative burdens. This study addresses these challenges by presenting a novel, replicable framework for translating complex institutional regulations into an automated, multi-stakeholder KPI management system. We detail the design and implementation of a web-based platform at Astana IT University, which was developed by programmatically encoding the institution's official KPI calculation and validation rules. The system features a multi-perspective analytical ecosystem, providing role-specific dashboards for faculty, review committees, department heads, and central administration to support synchronized decision-making. The core scientific contribution is a holistic methodology that combines stakeholder-driven requirements analysis with a "Policy-as-Code" approach to create a transparent, auditable, and scalable solution. Preliminary results indicate significant improvements in efficiency and data accuracy, demonstrating the framework's effectiveness. This study contributes not only a practical solution for KPI management but also a validated methodological blueprint for digital transformation applicable to other higher education institutions facing similar regulatory and administrative complexities.
Future work will explore the integration of predictive analytics to enable early intervention in cases of underperformance. Additional modules such as goal-setting tools, peer comparison features, and customizable reporting templates are also planned to enhance usability and strategic planning capabilities. By fostering a data-driven culture and ensuring alignment with institutional goals, such systems can play a key role in long-term academic quality assurance and workforce development
COMPARATIVE RESULTS OF USING DEEP LEARNING MODELS WITH ENSEMBLE METHODS FOR WILDFIRE ASSESSMENT
Get PDFWildfires are an increasingly transnational global environmental and socio-economic problem. In fact, their frequency, intensity and destructive power has grown drastically over the past decades largely driven by climate change, unsustainable land management and other human activities. Climate change has shown through rising global temperatures, longer and hotter droughts, and greater wind speeds, has fostered the perfect environment for fires to spark and sweep through the land. Kazakhstan is one of the Central Asian countries where the effects of climate change are making such disasters not only more frequent, but much worse. This vulnerability was tragically illustrated by the recent large-scale forest fire that swept across the Abay region, resulting in considerable ecological harm and exposing serious deficiencies in early detection and response capabilities. These advancements all point to an increasing, pressing need for more innovative, rapid, and dependable ways to evaluate, anticipate, and reduce wildfire risks. To address these issues, in this study we present a state-of-the-art ensemble-based deep learning approach to improve the accuracy and efficiency of wildfire detection. Our approach marries the strengths of two other state-of-the-art object detection algorithms, YOLO (You Only Look Once) and SSD (Single Shot Multibox Detector).By training this ensemble based model on a massive and varied dataset of landscape images and real-life wildfires, we’re able to get a general detection accuracy of 89%.This combined performance marks a striking advancement from when each model is used individually, especially in reducing false positives and providing more uniform and trustworthy results. Through fusing these models together and keeping them in one single unified framework there’s a notable boost in state-of-the-art detection accuracy as well as real-time image processing speed capabilities. This is a requirement for any real-time application. These results emphasize the value of using ensemble deep learning methods to enhance wildfire management and response strategies, eventually leading to more effective and proactive efforts
APPLYING DATA ANALYTICS AND BI SYSTEMS TO BUILD A STUDENT DIGITAL PROFILE: THE CASE OF ASTANA IT UNIVERSITY
Get PDFModern challenges of the digital transformation of education require the development of new approaches to assessing academic success and monitoring students' educational trajectories. This study presents a functional model of the data analytics system and visualization of the digital profile of a graduate of Astana IT University (AITU), based on the Integrated IGPA (Integrated Grade Point Average) indicator, which combines the academic, research, and social achievements of students. The aim of the work is to create a system of analytics, visualization, and interpretation of data reflecting the comprehensive development of students and their readiness for professional activity. The theoretical part examines modern approaches to educational analytics in higher education. A critical analysis of scientific sources, including research on learning analytics, educational data mining, and the formation of digital profiles of students, was carried out. The emphasis on technical aspects and insufficient connection with educational practice reveals the main limitations of the existing models.
The empirical part uses anonymized data from AITU students for 2022–2024, covering the indicators of Grade Point Average (GPA), Indicators of Research-Oriented Study (iROS), and Social Competition Indicators (SSCI). Dashboards built with the help of Power BI made it possible to visualize and interpret educational trajectories. The use of machine learning algorithms (K-means clustering, PCA analysis) ensured the typologization of student profiles. Using Python and the scikit-learn, seaborn, and pandas’ libraries allowed us to deeply explore the relationships between IGPA components.
The results of the study demonstrate the possibilities of personalized academic support, strategic management of educational processes, and increased transparency of student achievement. The developed model can serve as a basis for making managerial decisions and improving the quality of educational programs in the context of digital transformation.
The proposed approach can be scaled and adapted to other educational institutions, regardless of their size and specialization. Flexibility in integrating additional indicators reflecting the unique goals and values of a particular educational environment facilitates the model's versatility
MULTILINGUAL AUTOMATIC SPEECH RECOGNITION INTERFACE FOR TYPING: USABILITY STUDY AND PERFORMANCE EVALUATION FOR KAZAKH, RUSSIAN, AND ENGLISH
No full textWe present a multilingual automatic speech recognition (ASR) system for Kazakh, Russian, and English designed for the trilingual community of Kazakhstan. Although prior research has shown that speech-based text entry can outperform conventional keyboard typing for human–computer interaction and interaction with large language models (LLMs), little is known about its performance and usability in low-resource multilingual contexts, particularly for Kazakh. To address this gap, we fine-tuned a Whisper-based model on additional Kazakh speech data, achieving a large reduction in Kazakh word error rate (WER) from 21.55% with the OpenAI baseline to 8.84%, while preserving competitive performance for Russian and English. We then conducted a user study with 38 participants from Nazarbayev University, who performed dictated reading and editing tasks in all three languages. We evaluated performance using WPM, CPM, WER, and CER, and assessed usability and cognitive effort using the System Usability Scale (SUS) and the Raw NASA Task Load Index (NASA-TLX). Participants reached high speech-based typing speeds without editing and moderate speeds with editing across all three languages. Importantly, there were no statistically significant differences between Kazakh, Russian, and English in error rates, cognitive load, or perceived usability. Users reported low cognitive load (NASA-TLX < 40) and consistently high usability (SUS > 80%), indicating that the interface is efficient, easy to use, and requires minimal mental effort. These results demonstrate that Kazakh-adapted Whisper enables accurate, usable, and low-effort multilingual ASR, and highlight the potential of speech-driven text entry systems for trilingual contexts such as Kazakhstan