New Tool Bridges JSON and Semantic Web: AI Helps Scientists Convert Data to RDF Format

Scientists working with structured data face a persistent challenge: while JSON makes it easy to exchange information between systems, those systems often struggle to interpret the same data consistently. A new tool called MetaConfigurator aims to solve this by helping researchers convert their JSON, YAML, and CSV data into semantically rich RDF (Resource Description Framework) format—the foundation of the Semantic Web.

The Problem with JSON and Semantic Interoperability#

JSON and JSON Schema excel at structural validation, ensuring that data conforms to expected formats. However, they lack the semantic layer needed for machines to understand what the data actually means across different systems. This gap makes it difficult for researchers to share experiments, compare methodologies, or build knowledge systems that integrate data from multiple sources without manual interpretation work.

How MetaConfigurator Works#

The new RDF Authoring View extends the open-source MetaConfigurator JSON Schema editor with capabilities that let users transform raw data into ontology-compliant representations. The workflow includes:

• AI-assisted RML mappings: Users can define rules for converting JSON structures into RDF triples, with AI suggestions for ontology alignment
• Interactive triple refinement: Edit and validate semantic relationships directly within the interface
• SPARQL query generation: Natural language hints are converted into formal queries, making it easier to explore the resulting knowledge graph
• Knowledge graph visualization: Explore relationships and patterns in the transformed data
• Bidirectional sync: JSON-LD text views stay synchronized with RDF triple tables
• Ontology-aware auto-completion: IRI suggestions help users reference the right semantic concepts

Real-World Application: Laboratory Data#

The researchers demonstrated the tool using experimental protocols from metal-organic framework (MOF) synthesis—a chemistry domain where capturing precise procedural details, reagents, and quantities is critical. The workflow converts protocol JSON into ontology-based JSON-LD, then enables researchers to query connections between experimental conditions and their outcomes, all within a single integrated interface.