Research Programme
Metraltus investigates how the identity of food products can be documented and modelled across biological, technical, environmental, sensory, cultural and institutional dimensions. The programme is interdisciplinary by design: no single dataset or discipline can adequately represent a food system.
Core questions
- What makes a food product identifiable across origin, transformation and use?
- How can heterogeneous evidence remain traceable and open to revision?
- How can ontologies and knowledge graphs represent relations without erasing uncertainty, plurality or local knowledge?
- What conditions make computational methods useful and accountable in African agricultural and food contexts?
- How should scientific evidence, cultural meaning and institutional rules be connected without being confused?
Methodological framework
The programme combines critical literature review, terminological analysis, ontology engineering, knowledge representation, data-quality assessment, provenance modelling, responsible machine learning and scenario-based system design. Empirical validation will be reported only when a documented protocol, dataset, ethical basis and reproducible analysis are available.
Current workstreams
- computational representation of food identity;
- ontologies and knowledge graphs for food and agricultural data;
- trustworthy data pipelines and evidence provenance;
- artificial intelligence for sustainable and African agriculture;
- sensory, cultural and semiotic dimensions of food;
- food-system resilience in terrestrial and constrained environments.
Public boundary
This website presents the scientific purpose, vocabulary and public-interest orientation of the programme. It does not disclose proprietary implementation details, protected intellectual property or confidential research material.