11 September 2025 in IEEE Access
In the expanding frontier of natural language processing, the challenge of building accurate and semantically rich question answering (QA) systems for low-resource languages remains largely unresolved. This study presents a hybrid extractive QA model tailored for Kazakh, a morphologically complex and digitally underrepresented language, by integrating dense retrieval mechanisms with ontology-based semantic prefixing mechanism. Unlike conventional approaches that rely solely on retrieval and reading comprehension, our architecture injects dynamically constructed semantic definitions for domain-specific terms into the answer context, enabling more profound understanding and improved accuracy. Leveraging a novel dataset of Kazakh QA pairs generated through GPT-4 with expert validation, we introduce a dual-stream hybrid model (Hybrid B) that combines ontology-driven enrichment with fine-tuned transformer models for the retriever and reader components. The proposed system achieves a significant leap in performance, with 88% F1 and 76% exact match scores, substantially outperforming established baselines, including KazQAD. By anchoring the QA pipeline in a custom-built educational ontology relevant to the target domain, this research demonstrates how semantic structure can compensate for linguistic and data scarcity, paving the way for scalable QA systems in other low-resource languages such as Turkish. Beyond its empirical results, the study contributes a reproducible and modular framework for retrieval-augmented, ontology-aware QA, with implications for educational platforms, intelligent tutoring systems, and domain-specific information access. Through this work, we argue that the future of multilingual QA lies in hybrid architectures that combine symbolic structure with neural adaptability.
30 December 2024 in Bulletin of the National Engineering Academy of the Republic of Kazakhstan
LUL.N. Gumilyov Eurasian National University
L.N. Gumilyov Eurasian National University
Currently, ontologies are becoming increasingly popular among domain experts. They use formal descriptions of terms and the relationships between them to improve understanding of the subject area. Ontologies are widely used on the Internet – from large taxonomies categorizing websites to categorizations of goods sold and their characteristics. One of the most common systems for creating ontologies is the Protege program, as it provides a wide range of tools and functions that allow users to create, edit and visualize ontologies. This article explores the use of the Protégé ontology editor as a tool for building ontologies in computer science. The article begins with an introduction to ontologies and their meaning in computer science, followed by an overview of Protégé and its capabilities. The article also provides a step-by-step guide to creating a university ontology using Protégé, demonstrating the process of creating classes, properties and individuals, as well as the use of reasoning and query tools. The advantages and limitations of using Protégé for ontology development are also discussed, including its flexibility, extensibility and support for a wide range of plugins and tools. In conclusion, the article suggests possible directions for the future development of Protégé, including improvements to its user interface, integration with other tools and platforms, as well as expanded support for machine learning and natural language processing. In general, this article is a comprehensive introduction to the use of Protégé as a tool for building ontologies in computer science, offering practical recommendations and an understanding of the advantages and limitations of software. The article is suitable for researchers and practitioners in the field of computer science who are interested in developing ontologies and are looking for a convenient and powerful tool for this purpose.
30 June 2026 in Scientific Journal of Astana IT University
Reproducibility and comparability remain important methodological concerns in domain-specific question answering, particularly in low-resource educational settings. This study examines the impact of ontology-guided context filtering on extractive question answering in the Kazakh language under controlled and reproducible conditions. A benchmark corpus on the History of Kazakhstan was constructed from 11 school textbooks for grades 5–10 and canonically indexed using deterministic section identifiers. Fixed train–test splits, standardized chunking, and unified preprocessing and evaluation scripts ensured consistency across experiments. Three question answering systems were evaluated under identical constraints: a neural extractive baseline operating on unfiltered context, an ontology-only system relying on structured knowledge representations, and a hybrid system applying ontology-guided context filtering prior to neural answer extraction. Filtering was conducted under an inference setting that excludes access to gold document or section identifiers, preventing the use of privileged metadata. The hybrid system demonstrates modest improvements over the neural-only baseline. Exact Match accuracy increases from 68.87% to 70.45%, while the token-level F1 score improves from 80.02% to 81.68%. Ranking-based retrieval metrics also show small gains, with Mean Reciprocal Rank increasing from 0.870 to 0.892 and Recall@10 from 95.1% to 96.9%. Inference latency is reduced from 20.01 ms to 6.83 ms per question. Paired statistical testing confirms that the improvement in F1 is statistically significant, whereas improvements in Exact Match are less conclusive. Ablation experiments suggest that these differences are not attributable to context length reduction alone, but are associated with the interaction between ontology-guided context selection and neural reader training. Overall, ontology-guided filtering can provide incremental benefits for extractive question answering in a domain-specific, low-resource educational setting under transparent and reproducible experimental conditions.
29 July 2026 in Bulletin of Shakarim University. Technical Sciences
This study aims to explore and develop advanced methods for the effective detection of breast pathologies using state-of-the-art machine learning techniques, specifically YOLOv8 and Faster R-CNN. Traditional approaches to breast disease diagnosis are critically reviewed, and their effectiveness is evaluated in comparison to modern automated methods. The proposed models are applied to mammographic images to identify and categorize pathological patterns into six distinct levels, considering variations in severity and disease characteristics. This multi-level classification allows for a more precise assessment of disease progression and provides critical information for personalized treatment planning. Experimental results demonstrate that the proposed approach achieves high accuracy and fast image processing, enabling reliable and rapid detection of potential breast abnormalities. These findings suggest that machine learning algorithms can significantly enhance the diagnostic process, providing clinicians with more accurate and timely information. Furthermore, the study highlights the potential of automated detection systems to improve early diagnosis, optimize treatment strategies, and ultimately enhance patient outcomes. The results emphasize the growing role of artificial intelligence in medical imaging and its transformative impact on the future of breast disease management.
30 March 2026 in Вестник КазУТБ
Қазақ тілі үшін цифрлық ресурстар тапшылығы жағдайында Зияткерлік бағдарламаланатын жүйелерді қолдау үшін Қазақстан тарихы мектебінің курсының пәндік білімін ресімдеу ерекше өзекті болып отыр. Бұл жұмыста 5-10 сынып оқулықтарының толық корпусы негізінде құрылған Қазақстан тарихы бойынша зияткерлік бағдарламаланатын жүйелерді зерттеу нәтижелері ұсынылады және алынған білім графигінің QA-міндеттері үшін контекст іздеудің сапасына әсері талданады. Зерттеудің мақсаты-терминдерді теру алгоритмін әзірлеу (person, period, place, source, event, unknown). Бастапқы "астық" белгілерін алу үшін жоғары дәлдіктегі үлгілер пайдаланылады, содан кейін оқыту логистикалық регрессияны пайдалана отырып, XLM-R трансформаторының эмбеддингтерінде жүргізіледі. Seed-белгілеу бойынша валидацияда негізгі сыныптар (person, period, place) үшін 0,89 және F1-ден 0,92-ге дейінгі дәлдікке қол жеткізіледі. Зерттеудің ғылыми жаңалығы онтологияның практикалық пайдалылығын бағалау әдісін әзірлеу болып табылады QA-661 мысалдың жиынтығы (528 пойыз /133 тест) кепілдендірілген дұрыс жауап спандарымен. Ұсынылған әдіс бірнеше семантикалық іздеу конфигурацияларын салыстырады: оқулық бөлімдері бойынша негізгі іздеу, онтологиялық "құжаттар" бойынша іздеу және тезаурус пен онтология арқылы сұраныстар мен контексттерді кеңейту нұсқалары. Нәтижелер онтологиялық құжаттарды қолдану p@1-ді шамамен 0,56-дан 0,86-ға дейін және ndcg@10-ны ≈0,77-ден ≈0,92-ге дейін арттыратынын көрсетеді; біріктірілген "онтология + кеңейту" схемасы ndcg-нің жоғары деңгейін және сәйкес контексттерді толық қамтуды сақтайды. Осылайша, ұсынылған онтологиялық бағдарланған тезаурус пен Қазақстан тарихы бойынша білім графы 5–10 сыныптарға арналған оқу материалын құрылымдап қана қоймай, қазақ тіліндегі сұрақ-жауап жүйелерінде контекст іздеу сапасын айтарлықтай жақсартады және доменге бейімделген QA-модельдерді оқыту мен бағалауға арналған берік негіз қалыптастырады.