Refined range analysis of early warning and pre-control based on multi-element spatiotemporal clustering
2021
There are three parts in public security big data analysis when early warning and pre-control oriented: refined analysis scope, visualization of temporal and spatial characteristics of analysis scope, and subspace emergency analysis model for early warning and controlling. These three parts are interrelated, and complete the early warning and pre-control analysis of emergency iteratively. The exploration method of refined analysis range is analysed in the paper, which is based on the calculation and comparison of multi-element spatiotemporal clustering results. This research adopts a grid division strategy to convert heterogeneous data into a normalized space-time grid, in order to improve the versatility of the framework. And the estimation of grid records without observed values using kernel density estimation method. The added clustering ensemble is to improve the stability of clustering results. It can avoid the influence of uncertain factors on the clustering results, as well as improving the real-time performance of clustering algorithm, by the optimization of clustering ensemble parameters, and the clustering ensemble deployed on Hadoop platform.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI