Exploring Multi-Relational Crowd Flows Imputation with Distance-decay Effects

Junlei Yuan, China Population and development Research Center
Dan He, China Population and Development Research Center
Xuying Zhang, China Population and Development Research Center
Cuiling Zhang, China Population and development Research Center
Can Jia, China Population and Development Research Centre

The continuous crowd flows occurring in geographical space form stable mobility relationships within the space, constituting a spatial interaction network. Crowd flows imputation aims to grasp the patterns of crowd flows and predict the missing intensity of Crowd flows in geographical space. Graph neural networks excel in graph-structured interaction data. However, existing research often focuses on homogeneous networks, neglecting the impact of heterogeneous interaction relationships influenced by distance decay on flows imputation. Neglecting edge heterogeneity constrains the ability to effectively model the network structure, consequently leading to suboptimal performance in interaction imputation. This study introduces an interaction imputation graph convolutional network model. It constructs a heterogeneous interaction network with multi distance relationships, considering distance decay. The model performs graph embedding based on interaction relationships between nodes. It comprehensively incorporates multiple interaction modes, topological structures, and node attributes to enhance spatial interaction imputation accuracy. Empirically validated using , our model outperforms existing models, improving imputation accuracy by approximately 8.70%. Our model consistently maintains superior accuracy in interaction networks of various sizes, demonstrating the stable superiority of our model. We also demonstrated that a reasonable number of relationships and a larger feature dimension of geographical units yield better interaction imputation results.

Keywords: Population projections, forecasts, and estimations, Computational social science methods, Social network methods

See extended abstract.