The article focuses on enhancing stock market price prediction through artificial neural networks and machine learning. It underscores the significance of improving forecast accuracy by incorporating historical stock prices, macroeconomic indicators, news events, and technical indicators. Exploring deep learning principles, it delves into convolutional neural networks (CNN), recurrent neural networks (RNN), including long short-term memory (LSTM), and gated recurrent unit (GRU) modifications. This financial time series processing study covers data preprocessing, creating training/test sets, and selecting evaluation metrics. Results suggest promising applications for the developed forecasting models in stock markets, stressing the importance of considering various factors for precise forecasts in dynamic financial environments. Historical reserve data serves as the model foundation. Integration of macroeconomic, news, and technical indicators offers a holistic approach, aiding trend and anomaly identification for enhanced forecasts. The article recommends suitable deep learning architectures, highlighting LSTM and GRU's effectiveness in adapting to intricate data dependencies. Experimental outcomes showcase these architectures' benefits in predicting stock market prices, offering valuable insights for finance and asset management professionals in financial analysis and machine learning realms.

This study proposes a machine learning-based approach for automatic summarization of scientific documents using a fine-tuned DistilBART model a lightweight and efficient version of the bidirectional and auto-regressive transformers (BART) architecture. The model was trained on a large corpus of 12,540 scientific articles (2015–2023) collected from the arXiv repository, enabling it to effectively capture domain-specific terminology and structural patterns. The proposed pipeline integrates advanced text preprocessing techniques, including tokenization, stopword removal, and stemming, to enhance the quality of semantic representation. Experimental evaluation demonstrates that the fine-tuned DistilBART achieves high summarization performance, with ROUGE-2=0.472 and ROUGE-L=0.602, outperforming baseline transformer-based models. Unlike conventional approaches, the method shows strong applicability beyond academic research, including automated indexing of technical documentation, metadata extraction in digital libraries, and real-time text processing in embedded natural language processing (NLP) systems. The results highlight the potential of transformer-based summarization to accelerate scientific knowledge discovery and improve the efficiency of information retrieval across various domains.

The present study aimed to develop and evaluate a predictive approach for modeling pH dynamics during ayran fermentation using kinetic analysis, multivariate statistics, and IoT-based monitoring. Ayran is a low-viscosity fermented dairy beverage characterized by rapid acidification and high sensitivity to formulation and processing parameters, which limits the effectiveness of conventional endpoint pH control. The effects of two types of functional additives were investigated at empirically selected concentration ranges: a multifunctional dry fortifier (1–3%, w/w) and a vitamin–mineral syrup (2–4%, w/w). The data were analyed within a grouped comparative exploratory framework; the primary objective was comparative kinetic modeling of the fermentation process and prediction of the technological pH endpoint rather than evaluation of strict dose–response relationships. Fermentation was conducted at 41.0 ± 0.5 °C until the target pH range of 4.30 ± 0.05 was reached. An IoT monitoring architecture was used to generate a structured time-series dataset consisting of 909 sequential time-series measurements obtained during nine fermentation experiments. The formulation-associated kinetic behavior was evaluated using ANOVA, Tukey’s test, correlation analysis, and principal component analysis (PCA). The additives significantly affected the final pH and fermentation duration (p < 0.05; η² > 0.90). The application of additives reduced fermentation time by 1.6–1.8 h compared with the control. The first principal component explained 96.4% of the total variance, confirming that pH can serve as a sufficient proxy indicator for monitoring the overall fermentation system. A comparative analysis of four kinetic models (Gompertz, logistic, logarithmic, and linear) was performed over the full fermentation range (t = 0–10 h). The Gompertz model demonstrated the highest predictive accuracy (R² = 0.994–0.996; MAE = 0.027–0.033 pH units), whereas the logarithmic model was inadequate for describing the sigmoidal acidification profile of ayran (R² = 0.685–0.703). Numerical solution of the inverse problem enabled prediction of the time required to reach pH ≤ 4.35 in experimental groups with an accuracy of 0.90–1.93 h, providing a preliminary early warning signal suitable for operator decision support. For the control group, the asymptotic behavior of the Gompertz model during the stabilization stage limited the applicability of numerical prediction, indicating the necessity of direct pH sensor monitoring for this formulation. The proposed approach may serve as an exploratory basis for further development of predictive monitoring frameworks for fermented dairy production.

Статья описывает влияние пандемии COVID-19 на различные сферы жизни, включая образование, где произошел значительный сдвиг к дистанционному обучению. Автор рассматривает роль больших данных и искусственного интеллекта в улучшении образовательных процессов, особенно в условиях глобальных кризисов. Исследование посвящено применению ансамблевых моделей машинного обучения для анализа образовательных данных, собранных во время дистанционного обучения в период пандемии COVID-19. Рассматриваются методы KNN, Random Forest и XGBoost, которые были объединены в ансамблевую модель для улучшения точности предсказаний изменений активности учащихся. Результаты показали положительную динамику учебной активности школьников в условиях дистанционного обучения, с точностью модели до 87%. Образцы данных были получены в ходе онлайн-опроса 35950 учащихся средних школ Казахстана, включая городские и сельские школы. Модель также позволила выявить причины повышения активности учащихся.

This study presents a Markov chain-based metric called the Reliability Reserve (τ), designed to estimate the time available for operator intervention during the ayran fermentation process. This indicator can be integrated into a digital twin forecast management system. The fermentation process was obtained using a DTMC (discrete-time Markov chain) and divided into five states according to pH (S1–S5). Laboratory samples were prepared from premium-grade cow’s milk sourced from the Zher-Ana farm and divided into three experimental groups: Control (without additives), Opt1 (3% additive), and Opt2 (4% additive). A sequence of states was created for the three studied groups (Control, Opt1, and Opt2), and the transition states of the matrix were calculated. The Reliability Reserve quantifies how much time is left before the system transitions from the target state to the acidification state. For the first group, P₄₅ was 0.200, corresponding to τ = 26.9 min. The incorporation of functional additives increased P₄₅ to 0.250, reducing τ to 20.9 min and shortening the operator intervention window by approximately 6 min. Markov chains were constructed using 101 interpolated time points obtained from five experimental pH measurements for each group. The original pH values were recorded at 2, 4, 6, 8, and 10 h of fermentation and linearly interpolated with a step size of 0.1 h to improve temporal resolution. Model robustness was evaluated using sensitivity analysis (±0.05 pH boundary shifts) and bootstrap resampling (n = 1000, 95% confidence intervals). The concept of Reliability Reserve is a practical decision-making tool in real time. It offers an alternative to traditional reliability indicators such as MTTF and allows integration into digital twin-based control systems.