A two-stage ML pipeline using Random Forest for I2P detection (99.96% accuracy) and XGBoost for exfiltration classification (91.11% accuracy) on 184,548 network flows from the SafeSurf Darknet 2025 dataset.
Resilience of the Invisible Internet Project: A Computational Analysis
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
I2P peer selection exhibits no significant geographic homophily, with assortativity near zero and same-country links matching random expectations.
citing papers explorer
-
Detecting Data Exfiltration through I2P Anonymity Networks: A Two-Phase Machine Learning Approach
A two-stage ML pipeline using Random Forest for I2P detection (99.96% accuracy) and XGBoost for exfiltration classification (91.11% accuracy) on 184,548 network flows from the SafeSurf Darknet 2025 dataset.
-
Geographic Patterns in I2P Peer Selection: An Empirical Network Topology Analysis
I2P peer selection exhibits no significant geographic homophily, with assortativity near zero and same-country links matching random expectations.