Skip to content

Scientist Use Cases

This diagram maps researchers to the FCC subsystems they exercise when running reproducible experiments, reviewing literature, and federating knowledge across projects. Scientists enter through docs/for-scientists/ and lean on the RAGPipeline, DocumentChunker, KnowledgeGraph, FederatedKnowledgeGraph, and SemanticRetriever. Deterministic simulations plus event replay give FAIR-compliant reproducibility, and the open science persona pack supplies pre-built role templates.

The diagram below traces the scientist audience-to-subsystem mapping.

graph LR
    A((Scientist))
    UC1[/"FAIR workflow"/]
    UC2[/"Literature review"/]
    UC3[/"Reproducibility"/]
    UC4[/"Research methodology"/]
    UC5[/"Concept knowledge graph"/]
    UC6[/"Cross-project RAG"/]
    S1[RAGPipeline]
    S2[DocumentChunker]
    S3[SemanticRetriever]
    S4[KnowledgeGraph]
    S5[FederatedKnowledgeGraph]
    S6[CollaborationEngine]
    S7[Open science personas]
    A --> UC1
    A --> UC2
    A --> UC3
    A --> UC4
    A --> UC5
    A --> UC6
    UC1 --> S7
    UC1 --> S4
    UC2 --> S1
    UC2 --> S2
    UC2 --> S3
    UC3 --> S6
    UC3 --> S7
    UC4 --> S6
    UC5 --> S4
    UC6 --> S1
    UC6 --> S5

The FAIR Workflow task anchors the audience: findability, accessibility, interoperability, and reusability all map onto concrete subsystems (namespace registry, semantic retriever, RDF serializers, event replay). Literature Review Agents use the full RAG stack with one of six chunking strategies chosen to match the corpus structure.

Reproducibility is the strongest differentiator: scientists run deterministic scenarios, serialize every event, and replay the log to verify identical outputs. The Concept Knowledge Graph and Cross-Project RAG tasks extend the same machinery to federated corpora via FederatedKnowledgeGraph and the 11-ecosystem namespace registry.

Use-case detail

  • Use case 1 - FAIR workflow: Researcher aligns a study with FAIR principles; pulls open science persona pack and knowledge graph export; docs/for-scientists/fair-workflow.md.
  • Use case 2 - Literature review: Researcher runs persona-aware RAG over a paper corpus; docs/for-scientists/literature-review-agents.md.
  • Use case 3 - Reproducibility: Researcher replays a deterministic simulation and a serialized event log; docs/for-scientists/reproducibility.md.
  • Use case 4 - Research methodology: Researcher structures a multi-agent methodology with collaboration sessions; docs/for-scientists/research-methodology.md.
  • Use case 5 - Concept knowledge graph: Researcher builds a domain KnowledgeGraph and exports to OWL or SKOS; docs/for-scientists/experimental-design.md.
  • Use case 6 - Cross-project RAG: Researcher federates retrieval across ecosystem namespaces; hits FederatedKnowledgeGraph and the NamespaceRegistry.

See also

  • Source: src/fcc/rag/pipeline.py, src/fcc/rag/chunking.py, src/fcc/rag/retriever.py
  • Source: src/fcc/knowledge/graph.py, src/fcc/knowledge/federation.py
  • Related class diagram: ../class-diagrams/rag-pipeline.md
  • For audience tier: docs/for-scientists/index.md