SKSevara Dzhumagaliyevna Kurakbayeva

Sevara Dzhumagaliyevna Kurakbayeva

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Sevara Dzhumagaliyevna Kurakbayeva has a total of 30 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • An explainable AI approach based on XGBoost for leadership index prediction

    by Zh. Kemelbekova, A. Kozhabekova, Zh. Umarova, S. Kurakbayeva, Zh. Ashirbayeva

    This article presents a machine learning-based approach to predicting the leadership index based on students’ learning activity, digital skills, and the level of use of artificial intelligence tools. In the course of the study, a synthetic dataset of 110 observations was formed, taking into account the statistical characteristics typical of educational analytics. Based on this data, an XGBoost regression model based on the gradient boosting algorithm was developed. The effectiveness of the model was assessed using the coefficient of determination (R²), root mean square error (RMSE), and mean absolute error (MAE). The results obtained showed that the proposed model works with high accuracy and can effectively identify nonlinear and complex dependencies in the data. In the course of the study, factor importance, permutation analysis, and partial dependence methods were used to interpret the model results. The analysis results showed that the performance indicator is the main determinant in the formation of the leadership index, and the level of use of artificial intelligence tools was identified as a factor with an important but heterogeneous effect. The scientific novelty of the study is characterized by the provision of an integrated approach to predicting students’ complex competencies by combining educational data analysis and explanatory artificial intelligence methods. The proposed method can serve as a basis for developing decision-making support systems in the field of educational analytics, identifying risk groups, and forming personalized learning strategies.

  • Comparative analysis of machine learning models in predicting the popularity of social network content

    by B. E. Aidaulet, K. A. Utebayev, A. T. Kalbayeva, A. B. Imanbayeva, A. N. Zhaksanova, S. D. Kurakbayeva

    This study examines the effectiveness of machine learning models in predicting the engagement metric, which characterizes the popularity of social media content based on posts published on Facebook. Linear regression, Ridge, Lasso, and Random Forest models were utilized during the experiment. From a methodological standpoint, special attention was paid to the issue of data leakage; a modeling scenario was developed to eliminate data leakage by using only those variables known prior to publication. Furthermore, considering the pronounced positive asymmetry and heavy-tailed distribution of the dependent variable, the impact of the log1p transformation on the generalization ability of the models was investigated. The application of the log1p transformation enhanced the prediction quality of the Ridge regression, reduced bias, and increased the R² score. Additional validation based on time-based splitting proved the methodological correctness of the obtained conclusions.

  • Mathematical modeling and computer calculation of a woven composite material layer with local technological defects

    by S. D. Kurakbayeva, G. Y. Yermekbayeva, A. B. Imanbayeva, L.Z. Zholshiyeva, A. T. Kalbayeva, M. Yegenova

    This article looks at mathematical modeling and computer calculations for a layer of woven composite material that has some local technological defects. From the studies done, theres a mathematical model that describes how a woven composite layer deforms when it has curved fibers and those defects. With current computer tech, they came up with efficient algorithms to handle big sets of data on stress and strain states. Then a special software was made using Python and the Streamlit framework, and some calculations got run through it.

  • Open Access

    Computer simulation of transition modes in flow reactors considering the multistage and reactions non-perfectness

    by Aizhan Tazhikhanovna Kalbayeva, Zhanat Rysbayevna Umarova, Sevara Dzhumagaliyevna Kurakbayeva, Leyla Mukhamedjanovna Musabekova, Madamin A. Amandikov, Kuttybek E. Arystanbayev

    Due to the variety of reaction types and schemes in chemical-technological apparatuses, a general engineering methodology to assessing how the transient modes and reactions multi-stage act the kinetics in conditions of occurrence of moving reaction fronts in flow apparatuses has not yet been developed. The paper devotes to constructing the mathematical models for several important cases of the problem mentioned, namely: for theoretically study the kinetic dynamics of the conversion process in a three-stage chemical reaction with an autocatalytic first stage and the presence of a mass source of one of the components. An original mathematical model for describing the chemisorption dynamics at the initial stage of forming a moving reaction front in flow-through apparatuses has been developed. A special algorithm and numerical solution for the initial absorption period have been constructed, and appropriate computer simulation has been implemented. The significant influence of multistage on the formation and on stability types of stationary states has been established. Expressions to evaluate the characteristics of the emerging oscillatory modes have been obtained too. The results can be used to assess the influence of control parameters on the reaction front movement speed.

  • Analysis of opportunities and prospects of interfaces development using Pyqt and Qt Designer based on Python

    by M. Auezov South Kazakhstan University, G. K. Tagai, S. D. Kurakbayeva, Zh. Sh. Ashirbekova, A. T. Kalbayeva, T. A. Ryskul

    This article discusses the current field of programming with Qt Designer applications, which is increasing the importance of the program after using the characteristics due to changes in the demand for interfaces with a user-friendly menu. In connection with the increase in the number of devices and service Services, the demand for the need for abundant use of the necessary and appropriate interface additional programs was described. The article pays special attention to the creation of automation-oriented software for the performance characteristics of the enterprise and optimization problems in the reliability of its services. Various aspects and issues related to the use of the software are considered, which is revealed thanks to additional software solutions in cases where they are less effective with the limitations imposed by simple methods. The need to automate this process is emphasized, which makes it possible to get an idea of the variety of tools and software solutions available for automating the creation and management of program interfaces. Python can be used through the object – oriented programming language in many areas, from creating websites to preparing neural networks. If possible, the range of services is built-in and has a wide range of applications thanks to a set of external libraries. This allows you to create many electronic services in the use of certain libraries, such as websites, games, programs, neural networks and much more. In the course of our research, we will pay special attention to the Qt Designer program in the PyQt library and in it. With their help, we can create any necessary programs for various electronic services. One of the main reasons for using this built-in electronic application is the presence of many additional tools that simplify and speed up programming in Python, in addition to its versatility with easy writing. In this article, we will try to highlight the Python frameworks and libraries that every programmer should use by conducting a study.

  • Open Access

    Optimization and control algorithm for calculating separating membranes pore shapes

    by Zhanat Umarova, Sevara D. Kurakbayeva, Aizhan T. Kalbayeva, Pernekul A. Kozhabekova, Zhalgasbek D. Iztayev, Nabat Suieuova

    This work was aimed at researching and developing a computer model, as well as an optimization and control algorithm for calculating pore shapes for separating membranes in separation processes. Two types of flow fluid rates through separation membranes with various pores configurations were considered. Here, the first type of pores had the form of a tortuous channel with sharp narrowing endings, which, in turn, could be assumed as “zero” wall thickness holes. The second type of pores had conical narrowing at the ends, which is closer to the real shape of the pores. As a result of the analysis of the performed calculations and models, data were obtained on the effect of the thickness of the selective layer on the performance of the membrane. In computer calculations, it was determined that the optimal volumetric porosity should be limited to 0.15, which affects the mechanical strength of the membrane. Calculations were also carried out on the Chemcad software product, which resulted in graphs of the optimal permeability of various liquids, depending on the differ curvature and pore volume of the separation membranes.


  • Open Access

    Using the boundary elements approach to solve the problems of liquid filtration via the dam

    by A. S. Muratov, A. T. Kalbayeva, S. D. Kurakbayeva

    The work deals with the method and programs to calculate the liquid filtration via the dam, with allowing for the events of their possible damage. On the base of the boundary elements approach the method to calculate the flow of a liquid via the dam in the presence of a watertight fitting member was conducted. Using the numeric test, the free boundary position at the liquid filtration via the dam with the defected upper and lower basins was received and provisional analysis with the outcomes, received in the event of the defect of the upper dam basin only, was conducted. The primary free surface position has been determined arbitrarily. The boundary elements disposition drawing for the soil unit at the stream with the free area via the barrage with the defected upper (at the top) and lower barrage basins has been developed with allowing for reaching the optimal computing time. The boundary elements disposition drawing for the soil unit at the stream with the free area via the barrage with the defected upper (in the middle) and lower dam basins has been developed for the same considerations. The application was created in Delphi 7 visual programming environment. The main calculated outcomes were compared for the potential at the stream with the free area via the barrage with the defected only upper basin made earlier and for the barrage with the defected upper and lower basins. The outcomes show that in the event with the upper dam basin defect there is more wetting of the soil than in the event with the defect of both dam basins. It can be explained by the fact that the liquid stream at the defect lower dam basin will be faster and wetting of the soil will be less. The results of submitted paper can be useful in engineering practice for designing the soil dams.

  • MODELING OF DYNAMICAL REACTION-DIFFUSION SYSTEMS WITH MULTISTAGE AND NON-PERFECT KINETICS

    by Musabekova Leila, Information Technologies And Energy, Computing Systems And Software, Kalbayeva Aizhan, Dilman Viktor, Zhumataev Nurlybek, Kurakbayeva Sevara, Tauasarov Bolat