Article Per Year (5 Year)
p-Index From 2021 - 2026
1.006
P-Index
This Author published in this journals
All Journal Bulletin of Electrical Engineering and Informatics Buletin Ilmiah Sarjana Teknik Elektro Journal of Robotics and Control (JRC)
Serek, Azamat
Unknown Affiliation
Aerospace Engineering Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Mechanical Engineering

Published : 6 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 6 Documents
Search

 1 
Design of QazSL Sign Language Recognition System for Physically Impaired Individuals Zholshiyeva, Lazzat; Zhukabayeva, Tamara; Baumuratova, Dilaram; Serek, Azamat
Journal of Robotics and Control (JRC) Vol. 6 No. 1 (2025)
Publisher : Universitas Muhammadiyah Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.18196/jrc.v6i1.23879

Abstract

Automating real-time sign language translation through deep learning and machine learning techniques can greatly enhance communication between the deaf community and the wider public. This research investigates how these technologies can change the way individuals with speech impairments communicate. Despite advancements, developing accurate models for recognizing both static and dynamic gestures remains challenging due to variations in gesture speed and length, which affect the effectiveness of the models. We introduce a hybrid approach that merges machine learning and deep learning methods for sign language recognition. We provide new model for the recognition of Kazakh Sign Language (QazSL), employing five algorithms: Support Vector Machine (SVM), Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), Convolutional Neural Networks (CNN) with VGG19, ResNet-50, and YOLOv5. The models were trained on a QazSL dataset of more than 4,400 photos. Among the assessed models, the GRU attained the highest accuracy of 100%, followed closely by SVM and YOLOv5 at 99.98%, VGG19 at 98.87% for dynamic dactyls, LSTM at 85%, and ResNet-50 at 78.61%. These findings illustrate the comparative efficacy of each method in real-time gesture recognition. The results yield significant insights for enhancing sign language recognition systems, presenting possible advancements in accessibility and communication for those with hearing impairments.
Development of Method to Predict Career Choice of IT Students in Kazakhstan by Applying Machine Learning Methods Berlikozha, Bauyrzhan; Serek, Azamat; Zhukabayeva, Tamara; Zhamanov, Azamat; Dias, Oliver
Journal of Robotics and Control (JRC) Vol. 6 No. 1 (2025)
Publisher : Universitas Muhammadiyah Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.18196/jrc.v6i1.25558

Abstract

The growing intricacy of IT education requires resources to aid students in choosing specialized pathways. This study investigates the prediction of specialization preferences among IT students at SDU University in Kazakhstan through the application of machine learning techniques. The research contribution is the development of a predictive model that enhances academic advising by incorporating multiple factors, including academic performance, personality traits, qualifications, and extracurricular involvement. The research examined 692 anonymized student profiles and evaluated the efficacy of five machine learning algorithms: Random Forest, K-Nearest Neighbors, Support Vector Machine, Gradient Boosting, and Naive Bayes. Stratified 10-fold cross-validation was utilized to reduce the risk of overfitting. Gradient Boosting attained a peak accuracy of 99.10% in validation; however, its performance decreased to 92.16% on an independent test set, suggesting overfitting. Naive Bayes exhibited the lowest accuracy, recorded at 35.26%. Logistic regression analysis indicated a statistically significant correlation (p < 0.05) among academic performance, extracurricular involvement, and specialization selection. Personality traits and certifications significantly influenced the prediction process. The findings suggest that although Gradient Boosting demonstrates high effectiveness, the associated risk of overfitting requires additional refinement for practical application. The notable impact of academic performance and extracurricular activities indicates that educational institutions ought to prioritize these elements in student guidance. The incorporation of machine learning-based recommendations into advising frameworks enhances the precision of specialization predictions, thereby improving student decision-making and career alignment.
Resource-Efficient Sentiment Classification of App Reviews Using a CNN-BiLSTM Hybrid Model Baktibayev, Daulet; Serek, Azamat; Berlikozha, Bauyrzhan; Rustauletov, Babur
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 3 (2025): September
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i3.13954

Abstract

This study evaluates the performance of a hybrid convolutional neural network and bidirectional long short-term memory (CNN + BiLSTM) model for sentiment classification on user reviews from the Spotify mobile application. The primary aim is to explore whether competitive results can be achieved without relying on transformer-based architectures, which often require substantial computational resources. The proposed CNN + BiLSTM model combines local feature extraction with sequential context modeling and is benchmarked against traditional machine learning and simpler deep learning models, including a Random Forest classifier enhanced with polarity features, a standalone CNN, and a fully connected DNN. Sentiment labels were binary (positive or negative) and directly provided in the dataset without being inferred from star ratings. The dataset was balanced to avoid class skew. Experimental results indicate that the CNN + BiLSTM model achieves moderate improvements over the baseline models, with an accuracy of 0.8861 and an F1-score of 0.8691. While it does not surpass the highest-performing transformer-based methods reported in the literature, it performs comparably to several of them, despite having a lower computational footprint. Analyses of ROC curves, confusion matrices, and training dynamics further contextualize the model’s performance, showing strengths in classifying negative sentiments and convergence efficiency. To address overfitting, early stopping and dropout layers were employed as regularization techniques. The study contributes to the ongoing discourse on resource-efficient sentiment analysis by showing that hybrid architectures may offer a practical balance between model complexity and performance in specific application domains.
Few-shot brain tumor classification: meta- vs metric-learning comparison Akhmetzhanova, Shynar; Serek, Azamat; Kashayev, Ruslan; Kozhamuratova, Aizhan
Bulletin of Electrical Engineering and Informatics Vol 14, No 5: October 2025
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/eei.v14i5.10706

Abstract

Medical imaging requires accurate brain tumor recognition because precise classification is essential for early diagnosis and effective treatment planning. A major challenge in medical applications is that deep learning models typically require extensive amounts of labeled data to perform well. To address this, this research evaluates three few-shot learning (FSL) approaches-prototypical networks, Siamese networks, and model-agnostic meta-learning (MAML)-for brain tumor classification using the Figshare brain tumor dataset. The results show that prototypical networks consistently outperform the other approaches, achieving 89.07% accuracy (95% CI: 88.12–89.96%), 88.73% precision, and 88.67% recall, making them the optimal solution for this task. Siamese networks achieve 83.73% accuracy (95% CI: 82.64–84.76%), while MAML demonstrates significantly reduced performance, with 43.70% accuracy (95% CI: 42.10–45.22%). This study demonstrates that FSL can be applied effectively for medical image classification, with prototypical networks achieving the best performance in brain tumor detection. The inclusion of confidence intervals further validates the robustness and reliability of the results. Future research will focus on improving feature representation and exploring hybrid approaches to better handle rare tumor classes, thereby enhancing the clinical applicability of FSL models.
Accurate Crowd Counting Using an Enhanced LCDANet with Multi-Scale Attention Modules Abeuov, Nurmukhammed; Absatov, Daniyar; Mutaliyev, Yelnur; Serek, Azamat
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 3 (2025): September
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i3.14391

Abstract

Accurate crowd counting remains a challenging task due to occlusion, scale variation, and complex scene layouts. This study proposes ME-LCDANet, an enhanced deep learning framework built upon the LCDANet backbone, integrating multi-scale feature extraction via Micro Atrous Spatial Pyramid Pooling (MicroASPP) and attention refinement using CBAMLite modules. A preprocessing pipeline with Gaussian-based density maps, synchronized augmentations, and a dual-objective loss function combining density and count supervision supports effective training and generalization. Experimental evaluation on the ShanghaiTech Part B dataset demonstrates a Mean Absolute Error (MAE) of 11.50 (95% CI: 10.20–12.91) and a Root Mean Squared Error (RMSE) of 11.54 (95% CI: 10.26–12.99). Training dynamics indicate steadily declining loss and reduced validation MAE, while gradient norm analysis suggests reliable convergence. Comparative results show that, although CSRNet and SaNet achieve slightly lower MAE, ME-LCDANet attains a notably reduced RMSE, reflecting robustness against large prediction deviations. While the study focuses on a single benchmark dataset, the proposed architecture offers a promising approach for robust crowd counting in diverse scenarios.
Optimizing Event Replay with Hybrid Snapshotting in Event-Sourced Systems Shyngys, Akhulbay; Serek, Azamat; Kuanyshbay, Darkhan
Buletin Ilmiah Sarjana Teknik Elektro Vol. 8 No. 2 (2026): April
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v8i2.14214

Abstract

This study presents a hybrid framework for snapshotting and event replay optimization aimed at improving performance and scalability in event-sourced systems, where state reconstruction frequently experiences significant latency due to the replay of extensive event streams. Four snapshotting strategies were used to test this method: No Snapshotting, Fixed-Interval, Adaptive, and the new Hybrid method. The Adaptive strategy used dynamic triggers to make snapshots when an aggregate had more than 1,000 accesses in a five-minute window, when more than fifteen minutes had passed since the last snapshot, or when its event stream had more than 5,000 events. The Hybrid method used these adaptive triggers along with a fixed base interval of 5,000 events. For hot aggregates, the interval was cut in half to balance systematic coverage with runtime responsiveness. We built a prototype system in Java with Spring Boot and PostgreSQL. We used Kryo for snapshot serialization and Kafka to make controlled workloads. We ran tests on a 16-core Intel Xeon server to see how well it could handle event distributions that were uniform, skewed (Zipfian), or bursty, with anywhere from 10,000 to 1,000,000 events. We ran each configuration ten times and used 95% confidence intervals to find the average of the results. The results show that the Hybrid strategy consistently cut recovery latency, getting it down to 40 seconds under burst loads. This is a 58% improvement over Fixed-Interval and a 38% improvement over Adaptive. Throughput and responsiveness during normal operation were not affected, which is important. These results show that Hybrid snapshotting strikes a good balance between speed of recovery and storage overhead. This makes it a good choice for real-time, event-sourced applications. In the future, we will look into automatic tuning of snapshot intervals and adaptive thresholds, as well as adding support for distributed deployments and complex event schemas.
Co-Authors Abeuov, Nurmukhammed Absatov, Daniyar Akhmetzhanova, Shynar Baktibayev, Daulet Baumuratova, Dilaram Berlikozha, Bauyrzhan Dias, Oliver Kashayev, Ruslan Kozhamuratova, Aizhan Kuanyshbay, Darkhan Mutaliyev, Yelnur Rustauletov, Babur Shyngys, Akhulbay Zhamanov, Azamat Zholshiyeva, Lazzat Zhukabayeva, Tamara