ARCHITECTURE OF AN INTELLIGENT DECISION SUPPORT SYSTEM BASED ON SEMANTIC MODEL MATCHING

Abstract

The article explores a promising direction in the development of decision support systems (DSS), addressing the growing need for intelligent tools capable of processing complex, heterogeneous, and dynamically changing data. In situations where traditional DSS approaches often lack the flexibility to adapt to specific domain nuances and fail to adequately account for data semantics and user intent, this study fills a critical gap by introducing a new architecture based on semantic model alignment.The aim of the article is to describe the architecture of an intelligent DSS that employs semantic technologies to provide more accurate, context-aware, and user-centered decision support.Methods. Analysis of scientific literature, generalization, and systematization.Scientific novelty. The scientific novelty of the article lies in the formalization of the architecture of an intelligent decision support system based on the principles of semantic alignment of heterogeneous models to enhance the efficiency of the decision- making process.Results. The main idea is the integration of ontologies and semantic web standards to ensure interoperability between heterogeneous data sources and to automate the process of matching user queries with relevant knowledge models. The proposed system architecture consists of several core components, including a repository of semantic models, an inference engine, a user interaction module, and interfaces for external data integration. Special attention is paid to the dynamic configuration of decision-making models based on criteria of semantic similarity and relevance.Conclusions. The study demonstrates that semantic alignment mechanisms significantly improve the accuracy of recommendations and reduce the time required for decision-making in complex environments. Practical application scenarios are described, confirming the system’s adaptability and scalability. The conclusions emphasize the potential of semantic technologies in building intelligent DSS and highlight the importance of further research aimed at improving reasoning algorithms, user adaptation mechanisms, and the development of domain-specific ontologies. The proposed approach supports the evolution of DSS toward more flexible, intelligent, and context-aware systems.