See STRATEGIC_PLAN.md for the implementation roadmap.
- Strategic Dataset v1: Available in
dataset/v1- Contains structured strategic planning data - Financial Risk Dataset v1: Available in
dataset/financial_v1- Large-scale open financial risk dataset with 1,100+ labeled objects (phishing domains and scam wallets)
The new financial risk dataset includes:
- 1,100 labeled objects: 600 phishing domains + 500 scam wallets
- Comprehensive features: Domain analysis, wallet behavior patterns, risk metrics
- Privacy-preserving: Pseudonymous identifiers with salted SHA256 hashing
- Multiple formats: JSON, CSV exports
- Benchmark results: 89.55% accuracy on risk classification
- Machine learning ready: Feature-engineered for ML model training
Quick start:
cd dataset/financial_v1/scripts
python generate_simple_dataset.py # Generate dataset
cd ../benchmark
python benchmark.py # Run ML benchmark