Skip to content

About This Project

Overview

This is a LinkML model for the QUDT (Quantities, Units, Dimensions and Types) ontology. QUDT is a comprehensive ontology that provides a structured vocabulary for describing physical quantities, units of measure, and their relationships.

Early Stage Project

This is a first draft attempt at converting the QUDT ontology to LinkML. It likely contains issues and limitations. In a very likely event you find a problem please file a GitHub Issue here. Or submit a pull request with your suggested improvement!

What is QUDT?

QUDT provides a comprehensive vocabulary for:

  • Quantity Kinds - Observable properties that can be measured (e.g., length, mass, temperature)
  • Units - Standard units of measurement (e.g., meter, kilogram, kelvin)
  • Dimensions - Fundamental dimensions (e.g., length, mass, time)
  • Relationships - Connections between quantities, units, and dimensions

What is LinkML?

LinkML is a modeling language for describing data structures and schemas. It provides:

  • Multiple serialization formats - Generate JSON Schema, Pydantic, Python, TypeScript, SQL, and more
  • Validation - Validate data against schemas
  • Documentation - Auto-generate comprehensive documentation
  • Interoperability - Bridge between different data formats and systems

Key Features

This LinkML model provides:

  • Class hierarchy from QUDT (so far only main ontology classes, additional ontologies and individuals are coming soon)
  • Generated artifacts in multiple formats (Python, JSON Schema, OWL, etc.)
  • Documentation with diagrams

Conversion Process

This LinkML model was created through an automated conversion process with the following steps:

  • Used ROBOT tool to convert the published QUDT ontology files to OWL Functional Syntax (OFN) format
  • Applied Schema Automator to generate the LinkML model from the OFN representation
  • Made manual corrections to address issues in the automated conversion (see below)

Manual Corrections Applied

Unfortunately, the automated conversion process did not yield a usable LinkML schema. The generated output had several structural and type issues. The following corrections were made to make the schema useful:

  • Replaced auto-generated types with LinkML standard types (e.g., string, uriList, stringList)
  • Added standard ontology prefixes (owl, rdf, rdfs, dc, dcterms, prov, voag, vaem, dtype)
  • Fixed issues with the PhysicalConstant class definition
  • General cleanup and normalization of schema structure
  • Removed redundant Thing mixin from Comment class (already directly inherited)

Known Issues & Limitations

This is a first draft with known limitations:

  • The automated conversion may not have captured all semantic nuances from the QUDT ontology
  • Some complex OWL patterns may not translate perfectly to LinkML
  • Not all QUDT constraints may be fully represented
  • The model may require additional validation rules
  • Some class hierarchies may need refinement
  • Most Dublin Core metadata attributes are missing from the LinkML model (could be added with an additional automated step)
  • Only classes from main ontology were transferred; individuals need to be transferred

Project Files

The project repository contains:

Contributing

We welcome contributions! This is an early draft and there's much room for improvement. Areas where help is particularly appreciated:

  • Improving the schema structure and class hierarchies
  • Adding better documentation and examples
  • Identifying and fixing conversion issues
  • Adding validation rules and constraints
  • Creating example data and use cases
  • Improving alignment with QUDT semantics
  • Testing and validation

Please open an issue or pull request if you have suggestions or improvements!

Resources

License

The QUDT ontology is licensed under Creative Commons Attribution 4.0 International License. Please refer to the QUDT license page for details.

This LinkML model derivative is provided as-is for community use and improvement.

Contact

For questions or suggestions about this LinkML model, please open an issue in the GitHub repository.

For questions about QUDT itself, please refer to the QUDT community resources.