Ethereum Fraud Detection: A comparative analysis of supervised learning approach

Ethereum is an application platform that distributes versions of intelligent contracts to thousands of people globally, utilizing blockchain to decentralize data. Ethereum is a global currency that is used to exchange value without requiring supervision or outside involvement. However, as e-commerce grows, the biggest threat to trade security is the proliferation of illegal activities like phishing, money laundering, and bribery. The need for strong defenses is highlighted by the inherent hazards of blockchain technology, such as the potential for fraud and cyberattacks. The stability and preservation of confidence in the financial system depend on an accessible network that is impervious to these kinds of assaults. Different classification algorithms, including Logistic Regression (LR), K-Nearest Neighbors (KNN), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) are being used in this work to detect Ethereum fraud. We are utilizing a dataset that includes rows of Ethereum cryptocurrency transactions along with rows of known unauthorized transactions. Notably, the “XGBoost” model identifies differences that might draw attention and avert possible problems in this task. Supervised learning for Ethereum fraud detection improves the ecosystem’s security and integrity.