Lawrencium-103/nigeria-public-health-dashboard

# Nigeria Public Health Dashboard AI-Powered Health Intelligence for 37 States Interactive maps, charts, voice-enabled chatbot, and NLP-driven data exploration — built for low-resource environments. --- ## Why This Exists Public health decision-making in Nigeria routinely stalls on one problem: **data is available, but not accessible to the people who need it.** A program officer in Sokoto should not need a data science team to answer "Which LGAs have the lowest immunization coverage?" A state epidemiologist should not wait three weeks for a BI team to pull zone-level malaria trends. A dashboard that requires expensive SaaS licenses, constant internet connectivity, and specialised training fails the people who need it most. This project is a reference implementation for what public health data infrastructure *should* look like in resource-constrained settings. It is not a prototype or a proof of concept. It is a working system with 1,665 curated data points, an NLP-driven chatbot that answers natural-language queries against the data model, and a voice interface that works without any cloud AI service. The design choices — static JSON data, zero external API dependencies, browser-native speech, progressive web app — are deliberate. Each one solves a specific constraint that public health organisations in Sub-Saharan Africa deal with daily. --- ## Architecture ``` ┌─────────────────────────────────────────────────┐ │ Browser (Client) │ │ │ │ ┌──────────┐ ┌──────────┐ ┌──────────────┐ │ │ │ Leaflet │ │ Recharts │ │ Web Speech │ │ │ │ Map │ │ Charts │ │ API │ │ │ │ (GeoJSON) │ │ (Bar/Line)│ │ (STT + TTS) │ │ │ └──────────┘ └──────────┘ └──────────────┘ │ │ │ │ │ │ │ └──────────────┴──────────────┘ │ │ │ │ │ DashboardClie …