Graph Node Scoring for the Analysis and Visualisation of Mobility Networks and Data
Urban mobility and geographical systems benefit significantly from a graph-based topology. To identify the network’s crucial zones in terms of connectivity or movement across the network, we implemented several centrality metrics on a particular type of spatial network, i.e., a Region Adjacency grap...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-09-01
|
| Series: | Urban Science |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2413-8851/8/4/155 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Urban mobility and geographical systems benefit significantly from a graph-based topology. To identify the network’s crucial zones in terms of connectivity or movement across the network, we implemented several centrality metrics on a particular type of spatial network, i.e., a Region Adjacency graph, using three geographical regions of different sizes to exhibit the scalability of conventional metrics. To boost the topological analysis of a network with geographical data, we discuss the eigendata centrality and implement it for the largest of our Region Adjacency graphs using available geographical information. For flow prediction data-driven models, we discuss the Deep Gravity model and utilise either its geographical input data or predicted flow values to implement an additional node score through the Perron vector of the transition probability matrix. The results show that the topological analysis of a spatial network can be significantly enhanced by including regional and mobility data for graphs of different scales, connectivity, and orientation properties. |
|---|---|
| ISSN: | 2413-8851 |