Abstract

Recently, graph neural networks have played a crucial role across various fields. In this paper, we design a Graph Convolutional Network (GCN) to analyze population movement at the city level. The model consists of four Graph Convolution (GC) layers, where each layer aggregates information from neighboring nodes and updates the feature representation of each city. We utilize population mobility data from China, which includes daily city-to-city movement information. The proposed GCN estimates the strength of relationships among all cities. Experimental results demonstrate that the model achieves improved performance in estimating city-to-city migration flow relationships.

Additional Comments