Manuk Manukyan, Associate Professor at the Chair of Information Systems of the YSU Information Technologies Educational and Research Center and a Candidate of Physical and Mathematical Sciences, participated online in the 18th International Symposium on Intelligent Distributed Computing (IDC 2026), held from October 12 to 14 in Thessaloniki, Greece.
At the symposium, Manukyan presented a research paper titled "An LLM-based Approach to Data Integration," which focuses on automating data integration using large language models. The paper was co-authored by Anna Malkhasyan, a graduate and lecturer at the center.
The researchers propose an approach designed to map data schemas from different sources to a unified Extensible Conceptual Model (XCM). The approach is based on an agent-based architecture that combines a large language model (LLM), an XCM-specific knowledge graph, concepts from the Extensible Conceptual Model and a structured system prompt.
Together, these components enable the automatic generation of mappings from source schemas to a target schema, expressed as an algebraic program. The researchers have also developed a prototype of the proposed approach, providing a basis for exploring the potential to automate schema mapping and data integration processes.
The joint research by Manuk Manukyan and Anna Malkhasyan demonstrates the potential of large language models to address data integration challenges, contributing to the development of modern data processing and management methods.
The International Symposium on Intelligent Distributed Computing provides an international platform for collaboration among researchers, lecturers and students in the fields of intelligent and distributed computing. Its scope covers artificial intelligence, data science, multi-agent systems, machine learning, cloud computing and distributed computing systems.
The symposium offers opportunities to present research findings, discuss recent technological developments and strengthen professional cooperation.
