A Survey on Blockchain Abnormal Transaction Detection

Blockchain technology has undergone rapid development in recent years, transactions on Blockchains, represented by prominent examples such as Bitcoin and Ethereum, are rapidly increasing in number. However, with the large volume of transactions, a variety of scams such as phishing and Ponzi schemes have become more prevalent and are often hidden among legitimate transactions. To combat these anomalies and fraudulent activities, it is necessary to adopt anomaly detection methods. In this article, we conduct an extensive survey of the current body of research in the area of blockchain anomaly transaction detection and analyze the current state of research by examining the key point involved in detection, including data imbalance, feature extraction, and classification algorithms. Additionally, we discuss the potential of applying graph convolutional networks (GCN) to the domain of blockchain anomaly detection and predict that GCN will likely become a mainstream approach in the near future.