Application And Research Of Network Data Intelligent Analysis Technology Based On Graph Neural Network

The Internet has become an indispensable part of people’s lives.People have gradually realized the importance of cyberspace security,and cyberspace governance has become an important part of social governance.Traditional cyberspace analysis methods require a lot of manual works and have been unable to solve in the large-scale net.How to use artificial intelligence to process network data will be the focus of cyberspace governance in the future.This paper proposes to introduce a network data analysis method based on graph neural network.This method uses the deep learning model to analyze the network data,which is efficient processing of network data.（1）Node classification method based on graph convolutional network with attention.In the anonymous network environment,different network participants have different roles in the network,such as authority nodes,entry nodes,etc.But the anonymity of the network makes it difficult to find out node types.This paper proposes a node classification model based on graph neural network,which can find out the type of nodes through the attribute characteristics of different nodes in the network and the topology characteristics of the network structure,so it can describe the situation in the anonymous network.（2）Link prediction method based on graph neural network.In an anonymous network,due to the existence of multi-hop forwarding and encryption in the network,it is difficult to get complete link information in the network,and the topology structure in the net cannot be obtained effectively.In this method,the model is used to simulate the traffic transfer in the network,which can effectively capture the characteristics of the network in the time domain.With the analyzes of outflow and inflow in each node,the topological structure transformation of the entire network is restored,which can effectively carry out cyberspace topology situation.（3）Design and develop an intelligent analysis system for cyberspace data.The system can visualize the situation in the anonymous network based on the link prediction model and node classification model.The system can meet the fast processing of large data scale,and has good scalability.It can be stable on domestic servers and operating systems.