Gleipnir/Nidus-iOS
7
0
Fork 0
Code Issues 2 Pull requests Projects Releases Packages Wiki Activity Actions

doc: document current knowledge graph data extraction model #2

Merged
eliribble merged 2 commits from issue-1-knowledge-graph-doc into main 2026-07-22 01:47:01 +00:00
Conversation 3 Commits 2 Files changed 1 +638
ned commented 2026-07-22 01:19:48 +00:00
Member
Copy link

Summary

This PR adds comprehensive design documentation for the Nidus-iOS Knowledge Graph — the on-device voice dictation extraction engine that structures field technician speech into typed observation records.

What's documented

  • Pipeline architecture — end-to-end data flow from SFSpeechRecognizer through NaturalLanguage tokenization to struct extraction and UI rendering
  • Natural Language Service — lexical class and lemma tagging using Apple's NaturalLanguage framework
  • Extraction engine — gram-based keyword dispatch, negation handling, numeric extraction, volume/dimension parsing, condition matching
  • Knowledge Graph data model — all 7 sub-graphs (Breeding, Source, Fieldseeker, AdultProduction, Driver, Facilitator, RootCause) with their fields, enums, and implication chains
  • AudioTag system — parallel keyword-based categorical tag matching (30+ habitats, 18 species, 34 products)
  • Report types and workflows — Inspection vs MosquitoSource reports, treatment recommendations, user tags
  • Transcript tag highlighting — multi-colored word-level annotation for live transcription display
  • Test coverage — 20+ unit tests with realistic field dictation input
  • Known limitations — gram brittleness, no session persistence, no multi-value support, volume assumptions
  • Comparison to backend — how the iOS knowledge graph relates to the nidus-sync backend parser (Issue #155)

Related

Closes Issue #1 — documents the current data extraction model and provides context for understanding what Benjamin was requesting in nidus-sync Issue #155.

## Summary This PR adds comprehensive design documentation for the Nidus-iOS Knowledge Graph — the on-device voice dictation extraction engine that structures field technician speech into typed observation records. ### What's documented - **Pipeline architecture** — end-to-end data flow from SFSpeechRecognizer through NaturalLanguage tokenization to struct extraction and UI rendering - **Natural Language Service** — lexical class and lemma tagging using Apple's NaturalLanguage framework - **Extraction engine** — gram-based keyword dispatch, negation handling, numeric extraction, volume/dimension parsing, condition matching - **Knowledge Graph data model** — all 7 sub-graphs (Breeding, Source, Fieldseeker, AdultProduction, Driver, Facilitator, RootCause) with their fields, enums, and implication chains - **AudioTag system** — parallel keyword-based categorical tag matching (30+ habitats, 18 species, 34 products) - **Report types and workflows** — Inspection vs MosquitoSource reports, treatment recommendations, user tags - **Transcript tag highlighting** — multi-colored word-level annotation for live transcription display - **Test coverage** — 20+ unit tests with realistic field dictation input - **Known limitations** — gram brittleness, no session persistence, no multi-value support, volume assumptions - **Comparison to backend** — how the iOS knowledge graph relates to the nidus-sync backend parser (Issue #155) ### Related Closes Issue #1 — documents the current data extraction model and provides context for understanding what Benjamin was requesting in nidus-sync Issue #155.
Add comprehensive design documentation for the Nidus-iOS Knowledge Graph
system — the on-device voice dictation extraction engine that structures
field technician speech into typed observation records.

This covers the full pipeline from SFSpeechRecognizer capture through
NaturalLanguage tokenization, gram-based extraction, the KnowledgeGraph
struct hierarchy (7 sub-graphs: Breeding, Source, Fieldseeker, AdultProduction,
Driver, Facilitator, RootCause), the AudioTag keyword system, UI rendering,
test coverage, known limitations, and the relationship to the backend parser.

Issue: #1
eliribble commented 2026-07-22 01:41:29 +00:00
Owner
Copy link

Please drop the "11. Relationship to Backend Extraction (nidus-sync)" section entirely. That backend extraction is going to be dropped.

Please drop the "11. Relationship to Backend Extraction (nidus-sync)" section entirely. That backend extraction is going to be dropped.
Eli requested dropping the 'Relationship to Backend Extraction (nidus-sync)' section entirely, since that backend extraction is being dropped.
ned commented 2026-07-22 01:42:16 +00:00
Author
Member
Copy link

Done — section 11 dropped in commit 99f5068, pushed to the branch. The PR now ends at the AudioTag limitations note.

Done — section 11 dropped in commit 99f5068, pushed to the branch. The PR now ends at the AudioTag limitations note.
eliribble approved these changes 2026-07-22 01:46:56 +00:00
eliribble deleted branch issue-1-knowledge-graph-doc 2026-07-22 01:47:01 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
eliribble
Labels
Clear labels
  
customer
Customers are aware of, and affected by the issue

  
prod
Issue has manifested in production

  
repro'd
Has been reproduced by engineering

  
retest
Issue is ready for retesting in staging

  
staging
Issue has manifested in staging

No labels
customer
prod
repro'd
retest
staging
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
Gleipnir/Nidus-iOS!2
No description provided.