Research on Node Classification and Link Prediction of Knowledge Graph in Complex Network Analysis

This paper discusses the application of knowledge graph in complex network analysis, especially for the two key tasks of node classification and link prediction. The framework of this paper combines the rich information in the knowledge graph with the topology of a complex network to capture deeper relationships between entities. In the aspect of node classification, a novel algorithm based on graph neural network is designed, which can effectively aggregate various features from knowledge graph, thus improving the performance and accuracy of the classifier. In the aspect of link prediction, the similarity calculation method combining entity attributes and relation information is introduced to predict the potential links in the network. In order to verify the validity and robustness of the proposed algorithm, the author conducts detailed experiments on several real-world data sets and conducts comprehensive model simulation. Experimental results show that the proposed method has achieved significant improvement in both node classification and link prediction tasks. In addition, the model simulation further proves the stability and reliability of the algorithm.