December 31 · Article · КазУТБ · Artificial Intelligence
One of the traditional methods for community detection in knowledge graphs is agglomerative clustering. Agglom-erative hierarchical clustering is a widely used type of hierarchical clustering for grouping objects based on their similarity. This method follows a bottom-up approach, beginning with each individual data point considered as an independent cluster, which are then continuously merged based on a similarity threshold between clusters. This paper focuses on the use of agglomerative clustering for analyzing skills extracted from job postings on an online recruitment platform. It describes the approach to data collection, processing, and subsequent clustering, providing an overview of linkage methods between clusters and examples of the application of various coefficients for quantitative assessment of cluster quality. An analysis of bilingual clusters in Russian and English is conducted, al-lowing for an evaluation of the versatility and adaptability of the proposed approach to analyzing the multilingual labor market in Kazakhstan. It was found that agglomerative clustering methods hold significant potential for identi-fying structured groups of skills, which can enhance the understanding of labor market trends and needs. The analysis of clusters formed in different languages confirmed the universality and adaptability of the proposed ap-proach to multilingual data.
ATAkylbek TokhmetovOpen access
December 31 · Article · Eurasian Journal of Gender Studies · Education
Access to education for girls is among the global challenges strongly linked to poverty reduction, economic development, and social equity. Despite progress in narrowing gender disparities, significant gaps persist in regions where economic and demographic pressures prevail. In Kazakhstan, these challenges are compounded by uneven access to education, income disparities, and demographic dynamics. The current study explored the socio-economic and demographic factors influencing poverty levels and access to education. Six hypotheses were developed to analyze relationships between demographic structures, household income and expenditures, and educational access focusing on girls. Multivariate and univariate analyses assessed their impact on poverty depth and severity, preschool and secondary education coverage, and enrollment rates. The findings revealed that boys aged 0–14 significantly influenced poverty levels, while girls in the same age group showed no comparable effect. This reflects societal norms prioritizing boys’ education due to their perceived role as breadwinners, while girls are often deprioritized. Household income was also confirmed as a key determinant of access to education, particularly at the primary level, which forms the foundation for future opportunities. The results highlight the necessity of establishing strong social values and institutions that promote gender equality and women’s empowerment. The absence of such an environment hinders progress toward achieving the goals of gender equality.
Elvira Nurekenova (Madiyarova)Open access
December 31 · Preprint · arXiv · Renewable Energy, Sustainability and the Environment
This study presents a comprehensive numerical analysis of a quantum-dot-engineered heterostructure, PbS:$Yb^{3+},Er^{3+}$/CuBiO, optimized for water splitting applications. Using density functional theory (DFT) coupled with machine learning, the study explores the electronic, optical, and catalytic properties of the material. The optimized PbS structure exhibited a direct bandgap of 1.191 eV, while co-doping with Yb and Er transitioned the material to a metallic state, enhancing charge carrier mobility and electron-hole separation. The final heterostructure displayed an indirect bandgap of 0.431 eV, favorable for visible-light absorption. Key findings include an internal electric field strength of 6.3 Debye, efficient charge transfer confirmed by Bader analysis, and strong optical absorption at 2.4 eV. Machine learning models, including DNN and LSTM, were employed to predict photon absorption rates, achieving mean squared errors as low as 0.0004. The synergistic effects of enhanced internal electric fields, optimized band structures, and strong photon absorption underscore the materials' potential for efficient hydrogen production. This study bridges advanced computational methods and machine learning, offering a framework for designing high-performance photocatalysts in renewable energy applications.
DWDennis WayoOpen access
