EL-Grande/COARbaitsV1

BAITS Notify Prototype is a COAR Notify pilot aligned with the UNESCO-CODATA DPs4Crises-Pilot's crisis-resilient livestock traceability in Botswana. It implements Announce, Request–Review, and Offer patterns over W3C LDN with KLD relevance ranking and UNESCO-CODATA DPs4Crises alignment. Built with FastAPI · Streamlit · SQLite · Docker Compose. # BAITS — Botswana Animal Identification and Traceability System > The COAR Bots-egov-agri prototype — event-driven livestock traceability with KLD relevance ranking and UNESCO-CODATA DPs4Crises alignment. --- ## Overview COAR Bots-egov-agri is a research prototype showing how COAR Notify — an open protocol built on W3C Linked Data Notifications (LDN) and Activity Streams 2.0 (AS2) — can serve as the notification backbone for a national livestock traceability system during disease outbreaks and other agricultural crises. **Scenario**: 5 farmers across Botswana districts × 5 animals each. An active FMD alert in the North-East District. A veterinary authority (VET) processing movement permit requests ordered by crisis relevance. --- ## Features - **COAR Notify patterns**: Announcement (animal registration), Request–Review (movement permits), Offer (ownership transfer) - **LDN-compliant inboxes**: `POST` (201 + Location), `GET` (LDP Container), individual notification dereference - **KLD Relevance Ranking**: Kullback–Leibler Divergence over 6 Boolean features scores each notification against a crisis prior — surfacing the most atypical events first in the VET Work Queue - **Streamlit dashboard**: Farmer view, VET Work Queue (KLD-ordered), KLD Analytics tab, Admin view - **Colour-coded badges**: Red (KLD > 0.30) / Amber (0.15–0.30) / Green (< 0.15) --- ## Architecture ``` COARbaitsV1/ ├── baits_coar/ │ ├── docker-compose.yml │ ├── cleanup_db.py # wipe DB and start fresh │ ├── data/ # shared SQLite volume (baits_system.db) │ ├── ldn_inbox/ # FastAPI LDN receiver + KLD engine │ │ ├── Dockerfile │ │ └── main.py │ └── dashboard/ # Streamlit UI │ ├── Dockerfile │ └── dashboard.py ├── BAITS_Documentation.tex # Full documentation (Overleaf-compatible) ├── BAITS_Documentation.docx # Full documentation (Word) └── BAITS_KLD_Proposal.docx # KLD ranking proposal doc …