Detecting double-spending attack in Blockchain Networks using k-nearest neighbor algorithm

In this research, we propose a novel method that uses the K-Nearest Neighbors algorithm to identify abnormalities in double-spending attacks. This approach is analyzing and detecting patterns in digital transactions that may indicate the presence of double-spending attacks in cryptocurrencies like Ethereum and Bitcoin. We illustrate the process of training the suggested system using both simulated anomalous data and actual transaction data. The outcomes show how well the algorithm detects these attempts while preserving a low error rate.