Climate Policy Theme Analysis in Africa
# Climate Policy Theme Analysis in Africa
A Corpus-Based NLP Framework for Thematic Analysis of English Climate Documents and Policies in Africa.
## π Project Overview
This repository analyzes the distribution of 15 climate policy themes across three governance levels in Africa:
- **National Climate Policies** (African countries)
- **UNFCCC/Paris Agreement Submissions**
- **African Union Documents**
## π Key Findings
- **Themes are empirically distinct** - Maximum correlation between themes is only 0.67, confirming all 15 categories measure unique policy dimensions
- **Governance level matters** - Different policy frameworks prioritize different themes (see visualizations below)
- **Natural policy clusters** emerged from correlation analysis, showing how themes tend to co-occur
## π Visualizations
### Theme Correlation Clusters
### Governance Level Comparison
### Policy Heatmaps
| National Policies | Paris Agreement | AU Documents |
|-------------------|-----------------|--------------|
| | | |
## π Getting Started
### Prerequisites
- Python 3.8+
- pip
### Installation
```bash
# Clone the repository
git clone
github.com
cd MIT808_Capstone
# Create and activate virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -e .
# 1. Export Excel sheets to CSV
python notebooks/01_export_excel_sheets.py
# 2. Generate heatmaps
python notebooks/02_generate_heatmaps.py
# 3. Run correlation analysis
python notebooks/03_theme_correlation_analysis.py
Project Structure
βββ data
β βββ processed/ <- Cleaned CSV files
βββ notebooks/ <- Analysis scripts (numbered order)
βββ reports/figures/ <- Generated visualizations
βββ Data_Analysis/ <- Source code
β βββ config.py <- Theme definitions
βββ requirements.txt <- Dependencies