Study on Unsupervised Community Detection Algorithm Based on GCN Model

At present feature information on non-Euclidean distance data can be effectively extracted by the graph convolutional neural network. This paper provides an unsupervised community detection algorithm based on graph convolutional network model for application. The algorithm first selects some nodes in the graph to add artificial flags to simulate the input signal, which can satisfy the requirements of the propagation rule of the graph convolutional network. Then the algorithm would pass the flags to the neighboring nodes through the modified graph convolutional network propagation rule. While comparing the different flags on the same node and getting the attribution result of the nodes, the community partitioning results can be obtained by optimizing the result. Finally, the real-world data is used to test this algorithm and to evaluate by comparing with some classic community detection algorithms. The experimental results show that the algorithm provided by the paper can achieve a better result completely and correctly. in different types of graph data sets.