OLIY TA’LIM MUASSASALARIDA GETEROGEN MA’LUMOTLARNI SUN’IY INTELLEKT YORDAMIDA TAHLIL QILISH: MATEMATIK MODEL VA MAHALLIY YONDASHUV

Abstract

This article develops a mathematical model for the analysis of geterogeneous data accumulated in Uzbek higher education institutions (HEIs) through the HEMIS information system, electronic document management platforms, Learning Management Systems (LMS), and video and audio services. Data sources are formalized through a triple of features, an integration operator is introduced, and formal evaluation metrics for artificial intelligence (AI) models are presented. A comparative analysis of machine learning models is performed. The article also substantiates the legal and technical limitations of using global large language models (LLMs) due to data sovereignty and personal data protection concerns, and justifies the advantages of a local approach.