14 research outputs found
Порівняння моделей обробки казахської мови для покращення результатів семантичного пошуку
No full textThe object of the study is the text classification and semantic search tailored to the unique linguistic features of the Kazakh language. The research addresses the challenge of improving the accuracy, relevance, and efficiency of semantic search.
This study focuses on improving semantic search for the Kazakh language by analyzing computational models tailored to its unique linguistic features, such as agglutinative morphology and rich inflectional systems. The research compares traditional rule-based approaches and advanced transformer architectures, including fine-tuned models like RoBERTa, for their ability to handle semantic nuances, contextual relationships, and user intent. The results reveal that fine-tuned transformer models achieved significant advancements, with the RoBERTa model attaining a Precision@10 of 89.4 %, a Mean Reciprocal Rank (MRR) of 85.6 %, and an F1-Score of 88.0 %. Additionally, the semantic search system developed in this study demonstrated a precision of 88.4 %, recall of 87.6 %, and an F1-score of 88.0 % on a domain-specific Kazakh dataset.
Key to these improvements were innovations in preprocessing pipelines, including custom tokenization and lemmatization tailored to Kazakh's agglutinative morphology, and the integration of contextual embeddings to resolve issues such as synonymy and homonymy. Computational efficiency was enhanced through resource optimization techniques, enabling the deployment of these advanced models in constrained environments. These findings underscore the potential of tailored transformer models to bridge the gap in semantic search capabilities for underrepresented languages like Kazakh, advancing the inclusivity of natural language processing technologiesОб'єктом дослідження є класифікація текстів і семантичний пошук з урахуванням унікальних мовних особливостей казахської мови. Дослідження спрямоване на підвищення точності, релевантності та ефективності семантичного пошуку.
Це дослідження зосереджено на вдосконаленні семантичного пошуку для казахської мови шляхом аналізу обчислювальних моделей, адаптованих до її унікальних лінгвістичних особливостей, таких як аглютинативна морфологія та багаті флективні системи. Дослідження порівнює традиційні підходи, засновані на правилах, і вдосконалені архітектури трансформаторів, включаючи точно налаштовані моделі, такі як RoBERTa, щодо їх здатності обробляти семантичні нюанси, контекстуальні зв’язки та наміри користувача. Результати показують, що точно налаштовані моделі трансформаторів досягли значних успіхів: модель RoBERTa досягла Precision@10 89,4 %, середнього взаємного рейтингу (СВР) 85,6 % і F1-Score 88,0 %. Крім того, система семантичного пошуку, розроблена в цьому дослідженні, продемонструвала точність 88,4 %, запам’ятовування 87,6 % і F1-оцінку 88,0 % на казахстанському наборі даних для конкретної області.
Ключем до цих удосконалень були інновації в конвеєрах попередньої обробки, включаючи спеціальну токенізацію та лемматизацію, адаптовану до аглютинативної морфології казахської мови, а також інтеграцію контекстних вбудовувань для вирішення таких проблем, як синонімія та омонімія. Ефективність обчислень була підвищена завдяки методам оптимізації ресурсів, що дозволило розгортати ці передові моделі в обмежених середовищах. Ці висновки підкреслюють потенціал адаптованих моделей трансформаторів для подолання розриву в можливостях семантичного пошуку для недостатньо представлених мов, таких як казахська, що сприяє інклюзивності технологій обробки природної мов
Розробка гібридної моделі CNN-RNN для покращеного розпізнавання динамічних жестів у казахській жестовій мові
No full textWith around 1 % of the population of the Republic of Kazakhstan being affected by hearing disabilities, Kazakh Sign Language holds great importance as a means of communication between citizens of the state. The limitations of tools for Kazakh Sign Language (KSL) create significant challenges for people with hearing impairments in education, employment, and daily interactions. This research addresses these challenges through the development of an automated recognition system for Kazakh Sign Language gestures, aiming to enhance accessibility and inclusivity of communication using artificial intelligence. The approach employs advanced machine learning techniques, including Convolutional Neural Networks (CNNs) for recognizing spatial gesture patterns and Recurrent Neural Networks (RNNs) for processing temporal sequences. By combining these methods, the system recognizes both hand gestures and facial expressions, providing a dual-stream model that surpasses single-stream gesture recognition systems focused solely on hand movements. A dedicated dataset was created using Mediapipe Holistic, an open-source tool that identifies 543 landmarks across hands, faces, and poses, effectively capturing the multifaceted nature of sign language. The findings showed that the hybrid model significantly outperformed standalone CNN and RNN models, achieving up to 96 % accuracy. This demonstrates that integrating facial expressions with hand gestures greatly enhances the precision of sign language recognition. This system holds immense potential to improve inclusivity and accessibility in various settings across the Republic of Kazakhstan by facilitating communication for hearing-impaired individuals, paving the way for expanded research and application in other sign languagesДля близько 1 % населення Республіки Казахстан, яке має порушення слуху, Казахська жестова мова набуває великого значення як засіб спілкування між громадянами держави. Обмеження інструментів для Казахської жестової мови (КЖМ) створюють значні виклики для людей із порушеннями слуху в освіті, працевлаштуванні та повсякденному житті, обмежуючи їх можливості для повноцінної участі в суспільному житті. Це дослідження вирішує ці проблеми через розробку автоматизованої системи розпізнавання жестів Казахської жестової мови, спрямованої на підвищення доступності та інклюзивності спілкування за допомогою штучного інтелекту. Підхід використовує передові методи машинного навчання, зокрема згорткові нейронні мережі (CNN) для розпізнавання просторових моделей жестів і рекурентні нейронні мережі (RNN) для обробки часових послідовностей. Об'єднавши ці методи, система розпізнає як жести рук, так і вирази обличчя, забезпечуючи двопотокову модель, яка перевершує однопотокові системи розпізнавання жестів, зосереджені лише на рухах рук. Для дослідження було створено спеціальний датасет за допомогою Mediapipe Holistic — інструмента з відкритим кодом, який визначає 543 ключові точки на руках, обличчі та позах, точно і надійно охоплюючи багатогранну природу жестової мови. Результати показали, що гібридна модель значно перевершує окремі моделі CNN та RNN, досягаючи точності до 96 %. Це свідчить, що інтеграція виразів обличчя з жестами рук значно підвищує точність розпізнавання жестової мови. Ця система має величезний потенціал для покращення інклюзивності, доступності та якості спілкування в різних сферах Республіки Казахстан, сприяючи ефективному спілкуванню людей із порушеннями слуху та відкриваючи широкі можливості для розширення досліджень і застосувань в інших жестових мова
METHODS AND ALGORITHMS FOR SOLVING THE PROBLEM ON THE SUM OF SUBSETS
No full textWe study special-case algorithms for the subset-sum problem when the subset size is fixed to , using algebraic and geometric formulations that yield practical procedures with clear time and space bounds. The subset sum problem is one of the fundamental problems in computational complexity theory. It consists of determining whether, given a finite set of non-negative integers, there exists a subset whose sum of elements is equal to a predetermined number. This problem belongs to the class of nondeterministic polynomial time complete (NP-complete) problems: its solution can be verified in polynomial time, but an efficient algorithm for the general case has not yet been found. The goal of our research is to find new methods for solving the subset sum problem for special cases using algebraic and geometric approaches. The proposed method is based on a polynomial formulation of the problem inspired by Waring's conjecture for polynomials and the Neumann–Slater theorem. The main idea is to construct polynomials whose coefficients contain information about the sum of the elements of a subset. Using Vieta's theorem and the Euclidean algorithm, the problem is reduced to checking whether certain algebraic conditions are satisfied. The article proposes two lemmas proving the polynomial solvability of the subset sum problem for subset cardinality two and three. Based on them, two algorithms are developed: one uses value mapping and a fusion method, the other is based on a geometric criterion for collinearity of points obtained by transforming set elements. The algorithms demonstrate efficiency in terms of time and memory and do not require division into verification and decision stages. Effective methods for solving it allow us to develop faster algorithms for intelligent information processing, optimization of computing processes, and construction of reliable data protection systems. Our results establish polynomial-time solvability only for these fixed-???? cases and do not claim consequences for the general subset-sum problem or for the P vs NP question
DEVELOPMENT OF A SOUND-BASED MOBILE APPLICATION FOR ROAD ACCIDENT DETECTION USING MACHINE LEARNING AND SPECTROGRAM ANALYSIS
No full textRoad accidents continue to pose a serious threat to public safety, underscoring the need for innovative, automated emergency response systems. This study presents the development of a mobile application that detects road accidents by analyzing audio signals in real time and immediately sends SMS alerts with GPS coordinates to emergency services and user-specified contacts. The system comprises two parts: a user-facing Android application and a server-side component for data processing. To build and train the detection models, we leverage the MIVIA Road Audio Events dataset and applied preprocessing techniques including amplitude normalization, background noise filtering, and data augmentation. Feature extraction involved zero-crossing rate, spectral centroid, spectral flux, energy entropy, short-time Fourier transform (STFT), and Mel-frequency cepstral coefficients (MFCCs). Two classification approaches were investigated: traditional machine learning models (Support Vector Machine, Random Forest, Gradient Boosting) and a deep learning model based on convolutional neural networks (CNNs) using Mel spectrogram inputs. Experimental results demonstrate that the CNN model achieved the highest performance with 91.2% accuracy, 89.5% recall, and an F1-score of 90.3%, outperforming the best classical model (Random Forest), which achieved 85.1% accuracy. The system also reduced the average accident alert time from 5–7 minutes to 1–2 minutes, representing a 60–80% improvement in emergency response speed. These results confirm the system’s reliability and practical benefit, particularly in regions like Kazakhstan, where timely medical intervention is critical. Limitations include reliance on smartphone availability, internet access, and environmental sound conditions. Future work will explore real-world testing, integration of accelerometer and gyroscope data, and deployment of edge computing for faster on-device processing. Overall, the proposed solution is a cost-effective, scalable approach for improving road safety and saving lives through rapid, automated accident detection
ARTIFICIAL INTELLIGENCE-ENHANCED MOBILE DIAGNOSTICS USING DECISION TREES FOR EARLY DETECTION OF RESPIRATORY DISEASES
No full textThis 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
Principles of Foreign Language Teaching Taking Into Account the Discourse Analysis Technique
No full textThis consider uncovers the main standards actualizing the discourse investigation method using constructivism approach. The center of the consider is the research conducted by the author. The article provides an outline of the foremost vital components of discourse investigation, gives illustrations and methodological proposals. In conclusion, the conclusion is made almost the significance of interdisciplinary projects for the improvement of students' investigate aptitudes and a inventive approach to understanding issues
Expression and Purification of Viral Like Particles for Vaccines and Structural Analysis
No full textabstract: Succinylcholine-induced apnea is a common problem in pre-hospital medicine that affects 1/1800 patients who undergo rapid sequence intubation. Succinylcholine is an anesthetic that mimics the neurotransmitter, acetylcholine. It binds to cholinergic receptors, blocking acetylcholine access, and causes paralysis for (normally) only a short time. Butyrylcholinesterase, which is responsible for succinylcholine hydrolysis, is deficient in a small percentage of the population. Previous studies have shown that wild-type butyrylcholinesterase (BChE) can be produced in transient-expression Nicotiana benthamiana, and can reverse the effects of succinylcholine induced apnea through enzyme replacement therapy. The wild type enzyme is also capable of irreversibly binding and inactivating organophosphorus nerve agents and pesticides, and has also exhibited cocaine hydrolase activity. Super cocaine-hydrolyzing BChE mutants, which exceed 2000 times the catalytic capability of the wild-type, have been optimized and expressed in N. benthamiana. The purpose of this study was to determine whether these mutants also hydrolyze succinylcholine with improved efficiency. Variant 3 and Variant 4 exhibited catalytic efficiencies of 2.08 x 106 M-1 min-1 and 3.48 x 106 M-1 min-1, respectively, against their preferred substrate, butyrylthiocholine, in the Ellman assay. The wild-type plant-expressed BChE did exhibit hydrolysis of succinylcholine, as we had previously determined; however, neither Variant 3 nor Variant 4 demonstrated the ability to hydrolyze succinylcholine in our particular assay. Therefore, N. benthamiana-expressed Variant 3 and Variant 4 may not succeed as a dual treatment against cocaine toxicity and prolonged succinylcholine-induces paralysis
Plant-derived Virus-like Particles and Recombinant Immune Complexes as Potential Components of a Future HIV Vaccine
No full textabstract: HIV continues to remain a global health issue, in particular in many low and middle-income countries. The World Health Organization (WHO) estimates that of the nearly 38 million HIV-1 positive individuals, 25% are unaware they are infected. Despite decades of research, a safe and effective preventative vaccine has yet to be produced. The HIV-1 envelope glycoprotein41 and the Gag structural protein have been identified to be particularly important in HIV-1 transcytosis and cytotoxic lymphocyte response, respectively. Enveloped virus-like particles (VLPs) consisting of Gag and a deconstructed form of glycoprotein (dgp41) comprising the membrane proximal external region (MPER), transmembrane domain and cytoplasmic tail may present a unique and safe way of presenting these proteins in a state mimicking their natural formation. Another form of presenting the immunogenic glycoprotein41, particularly the MPER component, is by presenting it onto the N-terminal of an IgG molecule, thereby creating an IgG fusion molecule. In our lab, both VLPs and IgG fusion molecules are highly expressed and purified within GnGn Nicotiana benthamiana. The results indicated that these recombinant proteins can be assembled properly within plants and can elicit an immune response in mice. This provides a preliminary step in using such Gag/dpg41 VLPs and RIC as present a safe, effective, and inexpensive HIV vaccine
A critical review of dark tourism studies
No full textThe topic of dark tourism emerged in the last three decades as tourism became more accessible. It allows forgotten history to be revised and transferred to the public. This study aims to restructure existing categorization regarding dark tourism and address the research gaps in dark tourism studies. We collected studies from international publication databases – Scopus, Web of Science, and Google Scholar. We pre-processed the following data for each study: topic, authors’ location of university affiliation, study area, year of publication, top-cited articles, top productive journals in publishing dark tourism studies, keywords, and internality/externality of the author from the study area. With the current paper, we analysed review articles published from 1996 to 2024 (first quarter), applying qualitative methods. Based on these, a new analytical framework was generated. Furthermore, the connections between research topics were also analysed. The results of the analysis highlight specific research gaps in the literature on dark tourism and address poorly visible research fields in international journals, e.g. terrorism-related research, social media links of dark tourism, postcolonial contexts, or commemoration of communist past and heritage. Consequently, certain countries and regions are underrepresented in the literature. This critical review offers new research areas but also gives some directions to the theoretical enrichment of the dark tourism concept
