This study proposes an automated algorithm for constructing a punctuation-annotated dataset for training Transformer–BiLSTM models in Uzbek. The process includes text normalization, punctuation-aware tokenization, token-level labeling, dataset partitioning, and conversion into CoNLL format. A total of 11 labels are used, and the resulting corpus contains 11,888,213 tokens. The proposed method is intended for preparing large-scale annotated Uzbek data for punctuation restoration.
The article proposes a five-stage dialogue-level model for detecting social engineering attacks carried out through e-mail, SMS, messengers and voice calls. It was shown that existing tools analyse messages separately and therefore detect attacks that unfold as a multi-stage conversation too late. The model combines semantic features from a multilingual transformer, psychological persuasion principles and contextual features in a GRU layer, updates a risk score on a 0-100 scale after each new turn and warns the user before a harmful action occurs. Depending on the risk score, the conversation is passed, the user is warned or the conversation is blocked. Federated learning was used to train the model without centralising the confidential data of banks and mobile operators, and the SHAP method was used to produce understandable alerts. Proposals were made on organisational and technical countermeasures, evaluation of early detection performance and pilot testing at banks and operators.
This article is devoted to the direct determination of the coefficients of an interpolation polynomial in the monomial basis, proposing a new recursive method based on divided differences. The proposed method does not require an explicit expansion of the Newton interpolation polynomial into the monomial basis, nor does it require constructing and solving a Vandermonde system of linear equations. The method is based on the sequential recursive reconstruction of the divided-differences table. When a column of divided differences yields a constant value, this value is determined as the corresponding coefficient of the polynomial. The contributions of the coefficients found so far are then subtracted from the interpolation values, and the same process is repeated for the remaining lower-degree polynomial. As a result, all coefficients of the polynomial are determined sequentially, starting from the highest-degree term down to the constant term. The mathematical foundations of the proposed algorithm are rigorously proven through lemmas and theorems. In particular, the necessary and sufficient condition for the common value obtained in a column of recursive divided differences to be equal to the corresponding coefficient of the polynomial is established. The numerical example provided confirms the correctness of the proposed algorithm and demonstrates that the polynomial coefficients can be recursively determined with high accuracy based on the interpolation data. The proposed approach offers a new computational interpretation of divided differences and forms an alternative recursive computational scheme for determining polynomial coefficients.
The transition towards a green and sustainable economy is one of the important priorities of Uzbekistan’s long-term socio-economic development. The country faces interconnected challenges, including population growth, increasing energy demand, water scarcity, environmental degradation, and the need to maintain high rates of economic growth. This article analyzes the development trends of the green economy in Uzbekistan based on statistical data for 2020–2025 and national strategic targets set for 2030. Particular attention is given to the development of renewable energy sources, improving energy efficiency, reducing greenhouse gas emissions, rational use of water resources, and the relationship between economic growth and environmental sustainability. The findings indicate that Uzbekistan’s transition towards a green economy is closely associated with the development of the energy sector, investment attraction, technological modernization, and the improvement of environmental policy.
In medical diagnostics, we often encounter fuzzy, incomplete data and subjective clinical assessments, which can make traditional diagnostics unreliable. Intelligent decision-making systems based on fuzzy logic are an effective approach to overcoming such uncertainty in medical data. In this paper, a Mamdani-type fuzzy inference model is proposed for medical diagnostic decision-making. The proposed model represents clinical indicators using fuzzy sets, which allows processing of patients’ uncertain disease symptoms and medical parameters within the framework of rule-based inference. This approach includes identifying diagnostic parameters, creating a knowledge base from fuzzy data, fuzzifying input parameters, aggregating results, and defuzzifying to obtain interpretable diagnostic results. For example, in the paper, a diagnostic algorithm based on Mamdani fuzzy inference is developed and expressed using respiratory status indicators such as cough severity and shortness of breath. The proposed approach provides transparent and interpretable reasoning along with effective modelling of ambiguity in medical data. The results show that fuzzy inference systems can serve as a useful tool to support clinical diagnostics and increase the reliability of decision-making processes in intelligent medical systems.
