🤖 AI Summary
Geographic hotspot prediction suffers from weak high-dimensional spatial representation and low pattern recognition accuracy in existing methods. To address this, we propose a novel point cloud–voxel–community joint clustering paradigm: geographic events are modeled as 3D point clouds; spatial discretization is achieved via voxelization; a spatial similarity graph is constructed; and graph neural network–driven community detection is introduced to uncover multi-scale topological structural features. This approach pioneers point-cloud-driven voxel-level community partitioning, synergistically integrating geometric representation with topological analysis—thereby overcoming the modeling limitations of conventional grid- or density-based clustering for complex spatial patterns. Evaluated on the Turkey archaeological site dataset, our method achieves a 19.31% speedup over K-means and DBSCAN baselines, with only a 6% accuracy degradation, demonstrating superior efficiency and robustness.
📝 Abstract
Existing solutions to the hotspot prediction problem in the field of geographic information remain at a relatively preliminary stage. This study presents a novel approach for detecting and predicting geographical hotspots, utilizing point cloud-voxel-community partition clustering. By analyzing high-dimensional data, we represent spatial information through point clouds, which are then subdivided into multiple voxels to enhance analytical efficiency. Our method identifies spatial voxels with similar characteristics through community partitioning, thereby revealing underlying patterns in hotspot distributions. Experimental results indicate that when applied to a dataset of archaeological sites in Turkey, our approach achieves a 19.31% increase in processing speed, with an accuracy loss of merely 6%, outperforming traditional clustering methods. This method not only provides a fresh perspective for hotspot prediction but also serves as an effective tool for high-dimensional data analysis.