.png){kind=icon}
.png){kind=icon}
Return
Spatio-temporal graph data storage and calculation based on grid graph database
.png){kind=icon}
OA
AI
DOI：10.1080/17538947.2025.2531844.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
How to store and calculate knowledge graph data is an important research direction in database management. As the fundamental elements of spatio-temporal knowledge graph (STKG), spatio-temporal graph data are characterized by large amounts of data, heterogeneous types, and strong sparsity, making them difficult to effectively express in the widely used key-value databases and graph databases. This paper proposes a grid graph database (GGD) to store and manage spatio-temporal graph data. Through the Geographic Coordination Subdivision Grid with One-Dimensional Integrated Coding on a 2nTree (GeoSOT), GGD constructs a retrieval system for five types of child-tables, utilizing temporal and spatial grid codes to precisely identify and position the storage of spatio-temporal quadruplets. With the use of grid coding algebra, GGD can also calculate complex spatio-temporal relations and generate new quadruplets to answer dynamic spatio-temporal questions. Experimental results show that, compared with other databases, GGD has significantly higher query and calculation efficiency levels for spatio-temporal graph data, with an average improvement of 4–10 times. Moreover, spatially distributed parallel optimization-based strategy can further improve the speeds of large-scale spatio-temporal graph data calculations, confirming the high scalability and practicality of GGD.
Keywords:
Grid graph database
spatio-temporal graph data
spatio-temporal knowledge graph
grid coding algebra
spatially distributed parallel optimization
Journal
IF:
4.9
Papers:
1.9K
Citations:
4.7K