AI RESEARCH

Ontology-based knowledge graph infrastructure for interoperable atomistic simulation data

arXiv CS.AI

ArXi:2604.06230v1 Announce Type: cross The reuse of atomistic simulation data is often limited by heterogeneous formats, incomplete metadata, and a lack of standardized representations of workflows and provenance. Here we present an ontology-based infrastructure for representing and integrating atomistic simulation data as a knowledge graph. The approach combines domain ontologies with a software framework that enables data capture both from existing datasets and directly from simulation workflows at the point of generation.