EFFECTIVE FRAUD DETECTION IN BLOCKCHAIN USING XGBOOST WITH RANDOM FOREST

To solve the issue of fraud and abnormalities in the Bitcoin network, we offer a model. Here, we classify transactions based on integrated and fraudulent transaction patterns using machine learning methods like XGboost and Random Forest (RF). Then, future incoming transactions are predicted using the trained dataset. To identify fraudulent transactions, the authors use blockchain technology with machine learning algorithms. The proposed model performs a security analysis of the proposed smart contract to show the system’s robustness and determines the precision and AUC of the models to gauge accuracy. To guard against assaults and vulnerabilities on the suggested system, the authors additionally suggest an attacker model. Overall, as financial technology advances, the proposed approach seeks to offer a more secure method for spotting fraud in the Bitcoin network. The authors assert that their suggested system’s use of machine learning and blockchain technology may successfully identify and stop fraudulent transactions.