Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
PD ISO IEC TR 20943 — Procedures for Achieving Metadata Registry Content Consistency Technical Report Series (Metadata Mapping Procedure and Framework for Generating Ontologies)
Information Technology — Procedures for Achieving Metadata Registry Content Consistency: Metadata Mapping Procedure and Framework for Generating Ontologies (2013 Editions)
🗂️ PD ISO/IEC TR 20943 is the BSI-adopted ISO/IEC TR 20943 technical report series for information technology — procedures for achieving metadata registry content consistency. The two parts in this batch are Part 5, which specifies the metadata mapping procedure, and Part 6, which provides the framework for generating ontologies. Both are handled by BSI Technical Committee IST/40 (Data management and interchange). Working alongside ISO/IEC 11179 (metadata registry) and ISO/IEC 19763 (metamodel framework interoperability), the series provides methodological support for consistent semantic governance across systems and organisations, and is an important reference for data governance, data-exchange standardisation, and semantic interoperability.
💡 “The same-named data field has different semantics, and semantically identical data has different names” is the top pain point of data governance. Metadata mapping and ontology generation are the engineering methods that make such implicit semantic inconsistency explicit and manageable.
1. Series Composition and Framework 📌
| Standard | Title | ISO/IEC Edition | Core Content |
|---|---|---|---|
| PD ISO/IEC TR 20943-5:2013 | Information technology — Procedures for achieving metadata registry content consistency — Part 5: Metadata mapping procedure | ISO/IEC TR 20943-5 (Ed.1.0, 2013-11) | Metadata mapping procedure, mapping steps, and documentation |
| PD ISO/IEC TR 20943-6:2013 | Information technology — Procedures for achieving metadata registry content consistency — Part 6: Framework for generating ontologies | ISO/IEC TR 20943-6 (Ed.1.0, 2013-11) | Framework, method, and reference standards for generating ontologies from a metadata registry |
⚠️ This series is a technical report (TR) providing procedure and framework guidance, not mandatory data formats. Actual metadata registries should follow the structural requirements of ISO/IEC 11179, and ontology engineering may be implemented with existing standards such as W3C OWL/RDF.
2. Part 5: Metadata Mapping Procedure 🔄
Metadata mapping links metadata (data elements, data sets, etc.) from different sources and contexts, clarifying their semantic correspondence — a prerequisite for data integration and exchange. Key points of the Part 5 procedure include:
- Overview: purpose, scope, and participants of the mapping.
- Mapping steps: the full process from identifying source and target metadata, establishing correspondences, to recording mapping rationale and constraints.
- Documentation: results recorded in structured documents, including mapping type (one-to-one, one-to-many, etc.), dependency conditions, and semantic constraints.
- Consistency maintenance: how to keep source and target registry content consistent after mapping, avoiding drift as the registry evolves.
Through a regulated mapping procedure, organisations make implicit relationships such as “field A equals field B” explicit, providing a maintainable basis for data transformation, report consolidation, and system integration.
3. Part 6: Framework for Generating Ontologies 🧬
An ontology formally describes the semantics of concepts, attributes, and relationships — the advanced form of semantic interoperability. Part 6 provides the framework for generating an ontology from metadata registry content:
- Terminology basis: references definitions in ISO/IEC 11179-3 and ISO/IEC 19763-3, plus part-specific terms.
- Generation method: how to transform registered data elements, data classes, and relationships into ontology classes, properties, and axioms.
- Framework structure: phased generation workflow with quality control points.
- Linkage to existing standards: the dependence on and alignment with the metamodel framework (19763) and registry standard (11179).
✅ Moving from registry to ontology is reuse of semantic assets: already-governed data elements in the registry become the direct source of ontology classes and properties, avoiding “reinventing a terminology” and achieving “govern once, reuse many times”.
4. Summary of Key Technical Points 🔧
| Topic | Key Content | Document |
|---|---|---|
| Mapping procedure | Mapping steps, mapping types, documentation | 20943-5 |
| Consistency maintenance | Continued consistency of registry content after mapping | 20943-5 |
| Ontology method | Generating classes/properties/axioms from registry elements | 20943-6 |
| Terminology alignment | References to 11179-3 and 19763-3 definitions | 20943-6 |
| Dependent standards | ISO/IEC 11179, ISO/IEC 19763 | 20943-5 / -6 |
| Quality and reuse | Phased workflow, semantic asset reuse | 20943-6 |
5. Engineering Practice 🛠️
- 🔍 Data-exchange standardisation: before cross-system exchange, use the Part 5 mapping procedure to establish and document the semantic mapping of fields between parties.
- 🔧 Data governance platform: establish a metadata registry per ISO/IEC 11179 and generate a business ontology from the registry per Part 6 to support semantic search.
- 📊 System integration and migration: when merging or migrating systems, drive data-conversion rules from the mapping archive, reducing manual comparison.
- 📋 Knowledge graphs and semantic applications: use the generated ontology to build knowledge graphs, unified data dictionaries, and API semantic annotations.
6. Frequently Asked Questions ❓
❓ Q: What is the difference between a metadata registry and metadata mapping?
A: A metadata registry is a mechanism that centrally registers and manages metadata such as data elements and their attributes (ISO/IEC 11179); metadata mapping is the procedure for establishing semantic correspondences between metadata from different sources (Part 5 of this series). The registry provides the “standard vocabulary”; mapping solves “cross-vocabulary correspondence”.
A: A metadata registry is a mechanism that centrally registers and manages metadata such as data elements and their attributes (ISO/IEC 11179); metadata mapping is the procedure for establishing semantic correspondences between metadata from different sources (Part 5 of this series). The registry provides the “standard vocabulary”; mapping solves “cross-vocabulary correspondence”.
❓ Q: Why generate an ontology from a metadata registry?
A: Registry elements are governed and well-structured but limited in expressiveness; an ontology can express class hierarchies, property constraints, and inference rules. Generating an ontology from the registry reuses existing semantic assets, reducing ontology-building cost and ensuring terminology consistency.
A: Registry elements are governed and well-structured but limited in expressiveness; an ontology can express class hierarchies, property constraints, and inference rules. Generating an ontology from the registry reuses existing semantic assets, reducing ontology-building cost and ensuring terminology consistency.
❓ Q: What are ISO/IEC 11179 and 19763?
A: ISO/IEC 11179 is the metadata registry (MDR) standard covering identification, definition, classification, and registry management of data elements; ISO/IEC 19763 is the metamodel framework interoperability (MFI) standard covering registration and mapping of metamodels. Part 6 of this series references relevant clauses of both.
A: ISO/IEC 11179 is the metadata registry (MDR) standard covering identification, definition, classification, and registry management of data elements; ISO/IEC 19763 is the metamodel framework interoperability (MFI) standard covering registration and mapping of metamodels. Part 6 of this series references relevant clauses of both.
🔍 Q: What should a mapping document record?
A: It should record source/target metadata identifiers, mapping type (one-to-one/one-to-many/many-to-one), mapping direction, dependency conditions, semantic constraints, and the rationale and approval information — forming an auditable, maintainable mapping archive.
A: It should record source/target metadata identifiers, mapping type (one-to-one/one-to-many/many-to-one), mapping direction, dependency conditions, semantic constraints, and the rationale and approval information — forming an auditable, maintainable mapping archive.