Defense of the dissertation of Ramazanova Valiya for the degree of Doctor of Philosophy (PhD) in the educational program «8D06103 - Information systems»

L.N. Gumilyov Eurasian National University, a dissertation defense for the degree of Doctor of Philosophy (PhD) by Ramazanova Valiya on the topic «Application of knowledge graphs for information support of the educational process at the university» to the educational program «8D06103 – Information systems».

The dissertation was carried out at the «Information Systems education department» of L.N. Gumilyov Eurasian National University.

The language of defense is russian

Official reviewers:

Madina Mansurova – Candidate of Physical and Mathematical Sciences, Head of the Department of Artificial Intelligence and Big Data, Al-Farabi Kazakh National University (Almaty, Republic of Kazakhstan);

Dinara Kaibasova – Doctor of Philosophy (PhD), Associate Professor of the Department of Computer Engineering, Astana IT University (Astana, Republic of Kazakhstan).

Temporary members of the Dissertation Counci:

Ainur Kozbakova – PhD, Associate Professor, Leading Researcher at the Artificial Intelligence and Robotics Laboratory, Institute of Information and Computational Technologies (Almaty, Republic of Kazakhstan);

Akerke Akanova – Doctor of Philosophy (PhD), Associate Professor, Head of the Computer Science Educational Programs Group, S. Seifullin Kazakh AgroTechnical Research University NJSC (Astana, Republic of Kazakhstan);

Tatyana Batura – Candidate of Physical and Mathematical Sciences, Associate Professor, A.P. Ershov Institute of Informatics Systems SB RAS, Head of Laboratory (Novosibirsk, Russia).

Scientific advisors:

Madina Sambetbayeva – PhD, Associate Professor at the Department of Information Systems, L.N. Gumilyov Eurasian National University (Astana, Republic of Kazakhstan);

Iurii Zagorulko – Candidate of Technical Sciences, Head of the Artificial Intelligence Laboratory, A.P. Ershov Institute of Informatics Systems, Siberian Branch of the Russian Academy of Sciences (Novosibirsk, Russia).

The defense will take place on August 28, 2026, at 10:00 AM in the Dissertation Council for the training direction «8D061 – Information and communication technologies» in the educational program «8D06103 – Information systems» of L.N. Gumilyov Eurasian National University. The meeting of the Dissertation Council will be held in a mixed (offline and online) format.

Link: https://teams.microsoft.com/meet/47977030100942?p=I32iGtQjWp7mcayfwf 

Address: Astana, street Pushkina, 11, Educational building No. 2, room No. 222.

Abstract (English): ABSTRACT of the dissertation by Valiya Sapkireyevna Ramazanova, entitled «Application of knowledge graphs for information support of the educational process at the university», submitted for the degree of Doctor of Philosophy (PhD) in the specialty 8D06103 – Information Systems Purpose of the dissertation research. To develop a model and methods for applying knowledge graphs to build a recommendation system used as part of the information support of the educational process at a university, using the field of information technology as an example. Research objectives To analyze existing approaches to knowledge graph representation. To analyze the application of knowledge graphs in education. To analyze the data corpora required for constructing an integrated knowledge graph model. To develop an ontological knowledge graph model for integrating educational and professional competencies and skills. To evaluate and compare language models for generating vector representations of competencies and skills. To evaluate and compare clustering algorithms for forming groups of similar skills. To fine-tune, evaluate, and apply the selected language model to improve the integration and clustering of skills. To develop, train, evaluate, and compare recommendation models based on graph neural networks for knowledge graph recommendations. To train and evaluate an appropriate graph neural network-based recommendation model for recommending job vacancies and online courses. To develop a prototype recommendation system implementing the integrated knowledge graph model for recommending job vacancies and educational courses. Research methods. Methods of data collection and analysis, ontological design, experimental methods, graph theory, natural language processing, clustering, machine learning, graph neural networks, similarity metrics, correlation metrics, clustering quality evaluation metrics, recommendation quality evaluation metrics, and recommendation system development methods. Main propositions submitted for defense. A knowledge graph model has been proposed that integrates educational and professional data into a unified heterogeneous knowledge graph by combining educational competencies, professional skills extracted from job vacancies, and educational skills from online courses using multilingual sentence transformers based on semantic textual similarity. A model for educational recommendations based on a knowledge graph has been proposed using Heterogeneous Graph Transformers. A prototype information system for professional and educational recommendations has been developed, which can be used to improve the information support of the educational process at a university. Main research results. The first result is the proposed ontological model of a multidomain heterogeneous knowledge graph that provides semantic integration of university educational competencies, professional skills extracted from job vacancies, and skills from online courses. The integration is performed using a fine-tuned multilingual Sentence Transformer model and agglomerative hierarchical clustering. The second result is the proposed educational recommendation model based on heterogeneous graph neural networks (Heterogeneous Graph Transformers), which were applied for the first time to the task of link regression in a heterogeneous multidomain knowledge graph. The third result is the developed prototype of an information system for career and educational recommendations aimed at improving the information support of the educational process at a university. Justification of the novelty and significance of the research results. The scientific novelty of the research lies in the fact that, for the first time, an integrated knowledge graph model combining data from three domains (educational programs, labor market requirements, and MOOCs) has been proposed. In addition, heterogeneous graph neural networks have been adapted for educational recommendation tasks. Compliance with scientific development priorities and government programs. The research corresponds to the priority areas of digitalization and artificial intelligence development in the Republic of Kazakhstan, as well as the national policy aimed at improving higher education and training specialists for the digital economy. The study was carried out within the framework of the grant-funded project AP22783030 "Application of Machine Learning Algorithms for the Automated Adaptation of Educational Courses to Employer Requirements," confirming its compliance with the national priorities for the development of science and education. Description of the author's contribution to each publication. The research findings have been published in two papers in international peer-reviewed journals indexed in the Web of Science and Scopus databases (Q1–Q2), three papers in journals recommended by the Committee for Quality Assurance in Science and Higher Education of the Ministry of Science and Higher Education of the Republic of Kazakhstan, and presented in three papers at international conferences indexed in Scopus. The author's personal contribution includes the analysis of theoretical foundations and the development of practical solutions, the justification and selection of appropriate research methods, conducting experiments and evaluating their results, as well as designing and developing the knowledge graph model and the prototype recommendation system.

Dissertation

List of Scientific Works

Consultant's Review

Foreign Consultant's Review

Conclusion of the Research Ethics Committee