Recognize Illegal Transactions in the Bitcoin Network Using Graph Attention with DIKW

Abstract

Bitcoin has become the choice of many illegal transactions due to its anonymity. To fight crime and maintain the order of the financial market, it is necessary to identify illegal transaction activities in the Bitcoin network. On the basis of referring to the relationship between data, information, and knowledge in the DIKW architecture, this article proposes a new method called MP-GAT to convert discrete data in the Bitcoin network into usable information and knowledge to help identify illegal transactions. MP-GAT uses a combination of multi-layer perceptron and graph attention networks to build a model for identifying illegal transactions. Specifically, the concept of artificial intelligence is used to simulate the human reasoning system and learn from the data. Use the MP-GAT model to convert Bitcoin data into information and knowledge. The method of artificial intelligence is used to complete the transformation of data, information, and knowledge in the DIKW system to achieve the purpose of detecting and identifying illegal transactions in the Bitcoin network.