Cluster analysis of educational data for assessing and visualizing educational losses in school education: software implementation

- Pronina, Olha (orcid.org/0000-0001-7085-8027), Brodniuk, Оlena, Piatykop, Olena (orcid.org/0000-0002-7731-3051) and Fedosova, Irina (orcid.org/0000-0003-3923-8270) (2026) Cluster analysis of educational data for assessing and visualizing educational losses in school education: software implementation Information Technologies and Learning Tools, 3 (113). pp. 41-53. ISSN 2076-8184

Text Olha Pronina.pdf - Published Version Download (809kB)

Abstract

The article is devoted to the urgent problem of educational losses that arose as a result of the COVID-19 pandemic and military operations in Ukraine. The article considers the concept of educational losses, the prerequisites and causes of their occurrence, as well as the categories of losses. The experience of other scientists in measuring and compensating for educational losses is also analyzed. Losses in learning and the decline in the level of students' academic achievements are analyzed separately. The authors propose a comprehensive approach to measuring and compensating for educational losses using cluster analysis methods. The k-means and hierarchical clustering algorithms were applied, which made it possible to identify groups of students with similar academic achievements. Python, with the Pandas, Matplotlib, and Scikit-learn libraries, was used for data processing and analysis. The graphical interface was implemented to visualize the results and facilitate teachers' work. The article presents experimental studies on grouping students 5-6 grades into clusters, as well as loss calculations obtained using the developed software. The results confirmed the existence of a stable correlation between achievements in different subjects and made it possible to calculate the level of individual educational losses. The proposed approach creates conditions for personalization of learning, timely diagnosis of gaps and planning of compensatory measures, including differentiated tasks, group tutoring and additional psychological support. It was concluded that the automation of educational data analysis and the use of cluster analysis significantly increase the effectiveness of overcoming educational losses and contribute to ensuring equal access to quality education even in crisis conditions. This will allow teachers to better organize compensation for educational losses, adapt curricula to the individual needs of students.

Item Type:	Article
Keywords:	educational losses; distance learning; cluster analysis; individualization of learning; k-means; hierarchical clustering
Subjects:	Science and knowledge. Organization. Computer science. Information. Documentation. Librarianship. Institutions. Publications > 00 Prolegomena. Fundamentals of knowledge and culture. Propaedeutics > 004 Computer science and technology. Computing. Data processing > 004.4 Software Science and knowledge. Organization. Computer science. Information. Documentation. Librarianship. Institutions. Publications > 3 Social Sciences > 37 Education > 37.01/.09 Special auxiliary table for theory, principles, methods and organization of education > 37.02 General questions of didactics and method Science and knowledge. Organization. Computer science. Information. Documentation. Librarianship. Institutions. Publications > 3 Social Sciences > 37 Education > 37.01/.09 Special auxiliary table for theory, principles, methods and organization of education > 37.09 Organization of instruction Science and knowledge. Organization. Computer science. Information. Documentation. Librarianship. Institutions. Publications > 3 Social Sciences > 37 Education > 373 Kinds of school providing general education
Divisions:	Institute for Digitalisation of Education > Generic resouse
Depositing User:	Алла 1 Алла Почтарьова
Date Deposited:	03 Sep 2026 16:56
Last Modified:	03 Sep 2026 16:56
URI:	https://lib.iitta.gov.ua/id/eprint/750226

Downloads

Actions (login required)

View Item