Scientific Journal of Astana IT University
Not a member yet
250 research outputs found
Sort by
AUDIO-TO-TEXT TRANSLATION FOR THE HARD OF HEARING: A WHISPER MODEL-BASED STUDY
Get PDFThis study investigates the effectiveness of the Whisper model for audio-to-text transcription, specifically targeting the enhancement of accessibility for individuals with hearing impairments. The research focuses on the processing of audio recordings obtained from WhatsApp messenger, which often contain significant background noise that complicates speech recognition. To address this issue, advanced audio processing techniques were employed, including the use of the Librosa library and the Noisereduce package for noise reduction. The spectral gating methods applied in this study effectively diminished wind noise and other ambient sounds, allowing for clearer recognition of spoken content. To ensure the quality of the processed audio, we assessed its clarity using a SimpleRNN model. The training results demonstrated a progressive reduction in loss values across epochs, confirming the successful enhancement of audio quality. Once the audio files were adequately cleaned, we utilized the Whisper model, a sophisticated machine learning tool for speech recognition developed by OpenAI, to transcribe the audio into text. The transcription process yielded accurate Kazakh language output, despite the initial challenges posed by background noise. These findings underscore the critical role of high-quality audio input in achieving reliable transcription results and highlight the potential of machine learning technologies in improving communication access for hearing-impaired individuals. This study concludes with recommendations for future research, including the exploration of additional noise reduction techniques and the application of the Whisper model across various languages and dialects. Such advancements could significantly contribute to creating more inclusive digital environments and enhancing the overall user experience for individuals with hearing impairments
ASSESSMENT OF THE STATE OF DISRUPTIONS IN THE POWER SUPPLY SYSTEM OF A MOBILE COMMUNICATION BASE STATION
Get PDFThis study provides an in-depth analysis of power supply interruptions at mobile communication base stations (BS) operated by the Khorezm branch of Uzbekistan’s Uzmobile national mobile operator. The primary objective of this analysis is to evaluate the duration of power supply interruptions and their impact on the operational performance of base stations. In the case of the Khorezm region, data on power supply interruptions collected over one year from all districts were examined. According to statistical data, 13 base stations with the highest number of interruptions were selected for detailed analysis. The frequency, duration, and causes of these interruptions were studied to assess the reliability of the power supply system. The assessment results revealed distinct characteristics of power interruptions across different areas of the region. Special attention was given to evaluating the reliability of base stations from the perspective of power supply stability. The stability of the power supply system was used as the primary criterion in the analysis. The resilience of base stations to interruptions and the efficiency of their service were compared based on the frequency and duration of power outages. Additionally, the geographic location of the stations, the reliability of their connection to the electric grid, and other external factors were analyzed. The analysis identified significant differences in the intensity of power interruptions between districts in the Khorezm region. In some areas, the high frequency of interruptions was attributed to issues within the local energy infrastructure or natural conditions. Furthermore, the data enabled an assessment of how power supply interruptions affect the uninterrupted operation of base stations. This study draws important conclusions regarding the reliability of mobile communication infrastructure components, particularly base stations, in the Khorezm region. The findings emphasize the need for further research into eliminating energy supply issues, improving the efficiency of base stations, and enhancing the quality and continuity of communication services. The results of the analysis pave the way for developing technical and technological solutions to improve the reliability of base stations. Specifically, the implementation of alternative energy sources, such as supercapacitor banks or backup batteries, is recommended to provide rapid responses to power interruptions. Additionally, the advancement of monitoring and automated control systems is identified as an effective means to ensure the stability of base stations. This research serves as a crucial scientific and practical foundation for devising measures to improve the reliability of power supply systems in mobile communication networks
INTERFACE DESIGN OF AN INTELLIGENT INTERACTIVE LEARNING SYSTEM
No full textThis study presents the design, implementation, and evaluation of an Intelligent Interactive Learning System that employs multimodal interaction to improve the adaptability, accessibility, and engagement of digital education. Conventional e-learning platforms typically rely on static and text-based resources, which restrict personalization and reduce learner motivation. The proposed system integrates natural language processing, speech synthesis, and avatar-based interfaces to deliver lectures through synchronized speech, gestures, and facial expressions. The system automatically processes uploaded lecture scripts and slide presentations, segmenting and aligning them to generate interactive video lectures. A novel contribution of this work is the incorporation of customized Kazakh-language support, implemented through intonation modeling, dependency parsing, and gesture mapping to enhance inclusivity for underrepresented linguistic communities. The system performance was evaluated using Facebook’s variational inference text-to-speech model. Experimental results demonstrate real-time capability, with an average latency of 25.5 ms, throughput exceeding 4,200 characters per second, and low computational resource requirements. These findings confirm the suitability of the system for deployment in resource-constrained environments without compromising speech quality or responsiveness. Compared with conventional tutoring and static e-learning platforms, the system additionally provides automated assessment generation, multimodal feedback, and accessibility functions such as subtitles and adjustable playback controls. The study contributes a scalable model for intelligent, avatar-based learning that integrates speech synthesis, real-time interaction, and cultural-linguistic inclusivity. Future work will focus on extending personalization through adaptive learner modeling, incorporating affective computing for emotion-sensitive interaction, and enabling interoperability with established learning management systems
INTEGRATED APPLICATION OF MABAC, CODAS AND ARAS METHODS IN ASSESSING THE RELIABILITY OF INFORMATION SYSTEMS
Get PDFIn the modern digital world, the reliability of information systems has become one of the most important factors that determine the stability and efficiency of organizations. Even short-term system failures can cause serious financial losses, data breaches, and reputational damage. Therefore, assessing and improving the reliability of information systems is an essential part of ensuring their overall quality and resilience. To achieve an objective and comprehensive evaluation, this study applies multi-criteria decision-making (MCDM) methods that take into account both technical and organizational factors. The main focus is on the ARAS (Additive Ratio Assessment) method, which not only ranks the studied systems but also expresses the reliability level in percentage form. This makes the results clear, comparable, and easy to interpret in practice. For additional verification and comparison, the MABAC and CODAS methods were used. They help confirm the stability of rankings and support the validity of the conclusions drawn from the ARAS method. The selection of assessment criteria was based on the international standard ISO/IEC 25010:2023, which defines the quality model for software and information systems. Expert evaluations were carried out across ten characteristics — functionality, performance, compatibility, usability, reliability, security, maintainability, portability, recoverability, and adaptability. Using this data, all three MCDM methods were applied to calculate and compare the reliability of selected systems. The results show that ARAS provides a clear quantitative measure of reliability, while MABAC and CODAS strengthen the analysis by verifying ranking consistency. The combination of these approaches offers a practical and reliable framework for evaluating the quality and dependability of modern information systems
CORRELATION-MATRIX–DRIVEN DIAGNOSTICS OF INDUSTRIAL EMISSIONS: A PEARSON BASELINE WITH SCATTER-PLOT EVIDENCE
No full textCurrently environmental state became very actual in the world, especially in Kazakhstan. Air pollution of industries is a major threat to the environment and health of the people, especially in areas with high reliance on coal-powered power stations in electricity production. Fossil fuels in Kazakhstan are the largest electrical source, and they contribute to the emission of sulfur dioxide (S), nitrogen oxides (N), carbon monoxide (CO), and the particle matter (PM). Although, to formulate diagnostic and monitoring procedures at industry level it is crucial to determine relationships among emissions. The study approaches the Pearson correlation method on data taken from an automated emission monitoring system at the Coal Power Plant in Kazakhstan. The aim of the study is to discover linearity between emission indicators and industrial combustion. The observed correlation heat map and scatter-plots indicate positive trends among the CO and S, inverse correlation between CO and , and insufficient relation of CO and NO. These results show the key combustion processes, which involve reduced oxygen supply leading to the incomplete oxidation and simultaneous increased sulfur emissions. The three-dimensional description of CO dependence on S and further explains the coupled emission response and supports the explanation of underlying regularities in the operation. The correlation-based framework has diagnostic capabilities of the early identification of inefficient combustion regimes and enables scalable and data-driven methods of emission control. The research finds that Pearson-based analytics can be used to offer a strong and interpretable predictive modeling and regulatory monitoring foundation of future air-quality management in industries
BALANCING SPEED AND PERFORMANCE WITH LAYER FREEZING STRATEGIES FOR TRANSFORMER MODELS
Get PDFIn this paper, we evaluated different approaches to freezing BERT-base layers and analyzed their impact on the quality and speed of training in the task of named entity recognition in two languages. Layer freezing is an optimization technique in deep neural network training in which specific layers of a model remain fixed. This means their weights do not change during the backpropagation process. By not updating these layers, the overall number of parameters requiring adjustment is reduced, which results in lower computational demands and faster training times. Partial freezing of layers proved to be an effective way to preserve key representations of the model and ensure its adaptation to new tasks. Experimental results showed that freezing from three to six layers allows to achieve stable model performance regardless of the training language. Unlike standard approaches, our method highlights cross-linguistic applicability and promotes energy-efficient training. We personally designed the experimental setup, implemented the freezing scenarios, and carried out all performance evaluations. This study aims to evaluate the effectiveness of layer freezing in a pre-trained BERT model when performing the named entity recognition task. Two variants of the freezing strategy are considered: in the first one the upper layers of the model are fixed, in the second one the lower layers remain unchanged. The analysis is based on two corpora, the English language CoNLL 2003 and the Kazakh language KazNERD. Our experiments showed that freezing three to six layers provides the best balance between training speed and model quality. On the CoNLL-2003 dataset, the training time decreased from 266 to 167 seconds and the Macro F1 score remained at 87%. On KazNERD, learning accelerated from 1609 to 958 seconds with an accuracy of 94-95 % and Macro F1 in the range of 71-72 %. Full freezing of all 12 layers caused a dramatic drop in quality, with Macro F1 dropping to 50 % on CoNLL and to 7 % on KazNERD. This emphasises the importance of limited freezing and fine-tuning of the model architecture.
The study further examines how the choice of layers to freeze influences the model’s ability to adapt to new linguistic patterns and domain-specific terminology. These findings offer useful insights for researchers and practitioners aiming to enhance the efficiency of fine-tuning large language models while ensuring robust performance across different languages and datasets. The results also highlight the potential for optimizing resource usage in various NER applications without compromising critical language understanding
EDUCATIONAL FLOWS DISTRIBUTION SYSTEM OF UNIVERSITY ACADEMIC GROUPS
Get PDFPlanning of educational flows and academic groups by the Academic department is one of the responsible, complex and labor-intensive tasks solved at the stage of the educational process preparation in the university. When planning the work, the Academic department strives to improve the quality of work in order to achieve the best indicators for the types of educational work. One of the main tasks of the IT educational programs is the data analysis and construction of an automated system. Wide opportunities for the implementation of this goal are provided by the topic of scientific research, which is relevant both for the Academic department, which makes up the teaching load of the university, in which educational flows are involved, and for the university as a whole, for the teaching staff of all educational departments. The problem under consideration is really relevant in each higher educational institution and is of scientific interest due to the fact that insufficient attention is paid to mathematical modeling in software development. The aim of the paper is the development and software implementation of the automated system in Python to optimize the business process of distributing educational flows. For the efficiency of writing business processes for optimizing the formation of educational flows and academic groups, the architecture, algorithms and functional model of the software product are described. The functional model of the business process of forming educational flows and academic groups participating in the teaching load calculation at the stage of the educational process preparation at the university is considered. Moreover, the innovative product prototype has been created, which allows us to distribute educational flows evenly and quickly to a large extent, while fulfilling the accepted criteria and limitations of the model. The paper describes the architecture, algorithms and functional model of the software product, which corresponds to the IT field. The development of this innovative program will be useful both for beginners in Python programming and for developers creating their startups
INNOVATIVE APPROACHES TO AGRICULTURAL MARKETING: NSGA-II AND K-MEANS FOR STRATEGIES IN THE AGRO-INDUSTRIAL COMPLEX
Get PDFThe relevance of this study is determined by the urgent need to improve marketing strategies in the agro-industrial complex (AIC) of Kazakhstan, where competitiveness and sustainability depend not only on production efficiency but also on effective promotion of agricultural products. Rapid digitalization and regional market heterogeneity create new challenges for enterprises that cannot be solved by traditional heuristic or single-objective approaches. The purpose of the research is the development of a hybrid method for multi-criteria optimization of marketing budget allocation in the AIC of Kazakhstan. The objective of the experiment is to test the hypothesis that such a method provides more balanced solutions in terms of efficiency, coverage, and cost compared to baseline approaches. The methodology is based on combining the evolutionary NSGA-II algorithm with K-means clustering. The first stage identifies Pareto-optimal distributions of marketing resources, while the clustering procedure segments the obtained strategies into groups with distinct efficiency-cost trade-offs. Input data were derived from synthetic simulations reflecting typical market conditions and real indicators of several agricultural enterprises. The results of computational experiments demonstrate that the proposed method outperforms single-objective optimization. In particular, it achieved higher average efficiency (1.56 vs. 1.10), wider coverage (1.39 vs. 0.95), and greater hypervolume (0.67 vs. 0.45). Clusters with combined use of digital and television channels provided the most effective balance of performance indicators, while radio and print media remained relevant for enterprises with moderate budgets. The novelty of the study lies in integrating evolutionary optimization with machine learning for marketing strategy design in the AIC. The obtained data can be applied by managers and policymakers for media planning, budget allocation, and the development of adaptive strategies that strengthen competitiveness and contribute to export growth
AN INFORMATION TECHNOLOGY APPROACH TO PREDICT BREAST CANCER USING MACHINE LEARNING
No full textBreast cancer continues to be the most encountered malignancy in women globally and a leading cause of cancer-related mortality. This study describes an Information Technology approach to evaluate interpretable machine-learning methods for breast cancer prediction using routine clinical data and to situate performance against prior literature. All calculations are based on the Breast Cancer Wisconsin Diagnostic dataset (569 instances; malignant/benign labels) hosted by the UCI Machine Learning Repository. Each sample corresponds to a breast mass classified as malignant or benign. Four supervised machine learning models were applied: Logistic Regression with L1 penalty, Random Forest, Decision Tree, and Naïve Bayes, and compared the area under the ROC curve (AUC), accuracy, sensitivity, and specificity using DeLong’s test with Holm correction. The reproducible pipeline consisted of preprocessing, recursive feature elimination for feature selection, and a 5-fold cross-validation for hyperparameter tuning. Among the four models, the L1-penalized Logistic Regression yielded the best results, with an AUC indicating accuracy, sensitivity, and specificity of 99.6% (97.3%, 95.2%, 98.6%) on the test sets, respectively. This study illustrates the effective integration of supervised machine learning methods into diagnostic systems to produce early, accurate, interpretable diagnoses of disease. This study reinforces the proposed information technology approach for breast cancer prognosis. Limitations of the study are a moderately sized, homogeneous cohort, and restricted focus on structured variables, which may enhance internal validity while restricting generalizability. Our findings contribute to an emerging body of literature that well-tuned, regularized logistic regression provides a reasonable baseline against which breast cancer risk and other study outcomes can be compared, and a pragmatic route toward trustworthy AI in oncology
DEVELOPMENT OF A VIRTUAL REALITY-BASED FIRE SAFETY TRAINING SYSTEM FOR APARTMENT RESIDENTS
Get PDFFire safety training is crucial for preventing injuries and property damage during emergencies. Traditional training methods, such as physical drills and lectures, suffer from low engagement, high costs, and limited realism. Virtual reality technology offers an innovative approach to improve fire safety education by providing an immersive and interactive training environment. This paper presents the development of a virtual reality-based fire safety training system for apartment residents, designed for a head-mounted display. The system incorporates realistic fire simulations and mechanics for interactive fire extinguishing.
Our study explores both qualitative and quantitative measurements of a virtual reality (VR) training model, comparing it with alternative video-based training for fire safety. The experiment involved 20 participants who underwent training using either our VR system or video instructions, divided into two groups. After the VR training, participants completed a presence questionnaire and a knowledge test. Objective metrics included overall escape time and completion rate. Subjective data were collected through semi-structured interviews conducted after the experiments. Results indicate a significant difference in presence scores and higher knowledge scores for the VR group (VR: M = 10.9, SD = 2.37; Video: M = 7.3, SD = 1.33). These findings suggest that immersive VR training enhances procedural learning and situational awareness more effectively than passive video instruction. The study contributes to the field of VR in safety education by offering empirical evidence of its advantages and highlighting gaps in user engagement and realism in conventional methods. Limitations include the small sample size and short-term retention measurement. Future work will explore larger-scale evaluations and the integration of AR for blended learning experiences