SDG monitoring and analytics dashboard for East African governance and policy
# UN-SDG East Africa Dashboard Project
## 📌 Overview
This project builds a reproducible **data science + Power BI dashboard** to track progress on the United Nations Sustainable Development Goals (SDGs) across East African countries. It integrates open datasets from the UN, World Bank, WHO, and UNICEF, applies AI/ML forecasting, and delivers actionable insights for policymakers, NGOs, and donors.
## 🎯 Objectives
- Collect and harmonize official UN/World Bank indicators.
- Build a **star schema** for clean analytics.
- Apply **AI/ML models** (Prophet, anomaly detection, composite scoring).
- Visualize results in **Power BI** with interactive maps, KPIs, and forecasts.
- Provide transparent documentation and governance for reproducibility.
## 🌍 Scope
Countries covered: Kenya, Uganda, Tanzania, Rwanda, Burundi, South Sudan, and optionally DRC.
Indicators: Health (SDG 3), Education (SDG 4), Water & Sanitation (SDG 6), Climate (SDG 13).
## 🛠️ Tech Stack
- **Python** (pandas 2.3.3, pyarrow 22.0.0, scikit-learn 1.7.2, prophet 1.2.1)
- **Virtual Environment** (.ds-env) for reproducibility
- **Power BI Desktop** for visualization
- **GitHub** for version control & CI/CD
- **Open-source tools**: Prefect (pipelines), MkDocs (documentation)
## 📊 Key Insights
- Forecasts show which countries are **on track vs. off track** for 2030 SDG targets.
- Anomaly detection highlights **sudden declines** in food security or health.
- Composite SDG Index enables **prioritization of funding** and interventions.
- Dashboards connect SDG data to **UN challenges** like famine, conflict, and climate shocks.
## 📂 Project Structure
```
un-sdg-eastafrica/
├─ src/ # Python source code
│ ├─ ingest/ # Data ingestion scripts
│ │ ├─ pull_un_sdg.py # Fetch UN SDG Indicator data
│ │ └─ pull_worldbank.py # Fetch World Bank population data
│ ├─ transform/ # Data transformation scripts
│ │ ├─ harmonize_units.py # Standardi …