Fishing for Phishers: Learning-Based Phishing Detection in Ethereum Transactions

2025-04-24

AI Summary

This work addresses key challenges in on-chain phishing fraud detection on Ethereum: weak feature discriminability, severe class imbalance, and poor model generalizability. We systematically compare explicit transactional features against implicit graph-structural features, quantitatively revealing—for the first time—their complementary strengths and limitations in detection performance, temporal robustness, and adversarial adaptability. To address these issues, we propose a unified framework integrating graph neural networks (GNNs), imbalance-aware learning (SMOTE and Focal Loss), and SHAP-based interpretability analysis. Furthermore, we introduce a principled evaluation paradigm that jointly considers data composition and class distribution. Experiments demonstrate that graph features substantially improve cross-temporal generalization (F1 +12.6%), while the hybrid feature strategy achieves 98.3% precision and 92.7% recall on real-world streaming data. Our approach delivers a novel, interpretable, robust, and practically deployable solution for on-chain fraud detection.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Phishing detection on Ethereum has increasingly leveraged advanced machine learning techniques to identify fraudulent transactions. However, limited attention has been given to understanding the effectiveness of feature selection strategies and the role of graph-based models in enhancing detection accuracy. In this paper, we systematically examine these issues by analyzing and contrasting explicit transactional features and implicit graph-based features, both experimentally and analytically. We explore how different feature sets impact the performance of phishing detection models, particularly in the context of Ethereum's transactional network. Additionally, we address key challenges such as class imbalance and dataset composition and their influence on the robustness and precision of detection methods. Our findings demonstrate the advantages and limitations of each feature type, while also providing a clearer understanding of how feature affect model resilience and generalization in adversarial environments.

Problem

• Evaluating feature selection strategies for Ethereum phishing detection
• Assessing graph-based models' role in improving detection accuracy
• Addressing class imbalance and dataset composition challenges in detection

Innovation

• Combines transactional and graph-based features for detection
• Analyzes feature impact on model performance
• Addresses class imbalance and dataset challenges