ECtHR Citation Network Rankings Analysis

Analysis of centrality measures in European Court of Human Rights (ECtHR) citation networks.

Repository Structure

rankings/
├── scripts/              # Analysis scripts (documented in scripts/README.md)
├── docs/                 # Documentation  
├── results/             # Analysis results
│   ├── fixed-merged-subarticles-edges/    # Main centrality calculations
│   ├── high_low_analysis_50_cutoff/       # High/low performer analysis  
│   └── high_low_analysis_with_aggregates/ # Aggregate network analysis
├── networks/            # Network data (DO NOT MODIFY)
├── archive/             # Archived files
└── rankings.ipynb       # Main Jupyter notebook

Quick Start

Run Analysis

# Analyze high/low performers
python scripts/analyze_high_low_performers.py

# Test centrality combinations with verification
python scripts/test_combinations_with_verification.py

# Enhanced analysis with aggregates
python scripts/analyze_high_low_with_aggregates.py

Key Results

Location: results/fixed-merged-subarticles-edges/verification_test/

Best Combination: Degree + Eigenvector (43.2% win rate, 57/132 networks)

Documentation

• scripts/README.md - Complete script documentation
• docs/ - Methodology documentation
• results/high_low_analysis_with_aggregates/combination_analysis.txt - Combination justification

Data Directories (DO NOT MODIFY)

• networks/merged-article-edges/ - Network data
• results/merged-subarticles-edges/ - Legacy results
• results/pre/ - Pre-processing results