Orchestration of semantic analysis methods

Problem Statement. This paper examines a pressing issue: orchestrating semantic analysis methods to support data storage and navigation in the process of supporting scientific research. To address this issue, semantic analysis methods and the conditions for their applicability to various types of data are examined. Objective. To develop a semantic data infrastructure for intelligent processing of scientific information. Results. A multi-level ontology is proposed as a basis for representing scientific knowledge, along with a semantic scientific graph for knowledge navigation. Practical Relevance. The obtained results demonstrate the applicability of the proposed ontological approach for the integration and semantic analysis of scientific data, coupled with the use of large language models during knowledge graph navigation.