A Complex Network Node Clustering Algorithm Based on Graph Contrastive Learning

With the rapid development of complex network science, exploring the characteristics of nodes and their interrelationships in networks has emerged as a topical issue which has been extensively applied in a variety of scenarios, such as market analysis, social networks, and recommendation systems. In this paper, a complex network node clustering method based on graph contrastive learning is proposed in combination with a topology of the network and a behavioral analysis of the network nodes, which is used to deeply mine the preferences and behavioral patterns of the network nodes in order to formulate a differentiated recommendation strategy. The model automatically learns the deep feature representation of data by optimizing the distance relationship between positive and negative sample pairs, especially when dealing with complex and heterogeneous data, and is able to capture the underlying structure that is difficult to discover using traditional methods. Meanwhile, the model captures the global structure of the data by utilizing the correlation between data points and mapping the high-dimensional data to the low-dimensional space, which provides strong robustness and high clustering accuracy when dealing with non-linearly differentiable data. The research in this paper not only provides new ideas for clustering research in complex networks but also promotes the application of related methods of complex networks in multiple fields, which has important theoretical significance and practical value.