Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GeoAI-Enhanced Community Detection on Spatial Networks with Graph Deep Learning

Yunlei Liang, Jiawei Zhu, Wen Ye, Song Gao

TL;DR

The proposed GeoAI-enhanced unsupervised community detection methods called region2vec based on Graph Attention Networks (GAT) and Graph Convolutional Networks (GCN) perform the best when one wants to maximize node attribute similarity and spatial interaction intensity simultaneously within the spatial network communities.

Abstract

Spatial networks are useful for modeling geographic phenomena where spatial interaction plays an important role. To analyze the spatial networks and their internal structures, graph-based methods such as community detection have been widely used. Community detection aims to extract strongly connected components from the network and reveal the hidden relationships between nodes, but they usually do not involve the attribute information. To consider edge-based interactions and node attributes together, this study proposed a family of GeoAI-enhanced unsupervised community detection methods called region2vec based on Graph Attention Networks (GAT) and Graph Convolutional Networks (GCN). The region2vec methods generate node neural embeddings based on attribute similarity, geographic adjacency and spatial interactions, and then extract network communities based on node embeddings using agglomerative clustering. The proposed GeoAI-based methods are compared with multiple baselines and perform the best when one wants to maximize node attribute similarity and spatial interaction intensity simultaneously within the spatial network communities. It is further applied in the shortage area delineation problem in public health and demonstrates its promise in regionalization problems.

GeoAI-Enhanced Community Detection on Spatial Networks with Graph Deep Learning

TL;DR

The proposed GeoAI-enhanced unsupervised community detection methods called region2vec based on Graph Attention Networks (GAT) and Graph Convolutional Networks (GCN) perform the best when one wants to maximize node attribute similarity and spatial interaction intensity simultaneously within the spatial network communities.

Abstract

Spatial networks are useful for modeling geographic phenomena where spatial interaction plays an important role. To analyze the spatial networks and their internal structures, graph-based methods such as community detection have been widely used. Community detection aims to extract strongly connected components from the network and reveal the hidden relationships between nodes, but they usually do not involve the attribute information. To consider edge-based interactions and node attributes together, this study proposed a family of GeoAI-enhanced unsupervised community detection methods called region2vec based on Graph Attention Networks (GAT) and Graph Convolutional Networks (GCN). The region2vec methods generate node neural embeddings based on attribute similarity, geographic adjacency and spatial interactions, and then extract network communities based on node embeddings using agglomerative clustering. The proposed GeoAI-based methods are compared with multiple baselines and perform the best when one wants to maximize node attribute similarity and spatial interaction intensity simultaneously within the spatial network communities. It is further applied in the shortage area delineation problem in public health and demonstrates its promise in regionalization problems.

Paper Structure

This paper contains 31 sections, 13 equations, 5 figures, 3 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Community Detection Algorithms on Spatial Networks
  4. Graph Embedding
  5. Method
  6. Notations and Problem Definitions
  7. Algorithm
  8. Stage One: Node Representation Learning
  9. Stage two: Agglomerative Clustering
  10. Baseline Algorithms
  11. Louvain community detection
  12. Leiden community detection
  13. Random walk based model
  14. LINE
  15. K-Means
  16. ...and 16 more sections

Figures (5)

  • Figure 1: The workflow for community detection using the regions2vec method.
  • Figure 2: The 3D view of three metrics for all the methods. Left: the view of all methods, Right: the zoomed-in view without LINE
  • Figure 3: The distribution of all methods based on cosine similarity and intra-flow ratio.
  • Figure 4: The resulting communities maps of all methods.
  • Figure 5: The final Health Professional Shortage Area with scores for three methods.