用于犯罪空间聚集态研究的优化聚类算法
发布时间：2018-10-22 19:49
【摘要】：针对犯罪空间聚集态研究算法中有关窗宽优化选择问题,将动态优化窗宽算法与DENCLUE(DENsity-based CLUstEring)算法结合,提出一种新的聚类算法。以入室盗窃案件为例,研究了该算法在犯罪热点探测方面的应用。结果显示引入优化窗宽算法后,可得到较为精确的聚类中心位置及概率密度变化趋势,并且当格网边长与邻近格网距离阈值及邻近点距离阈值之比为2:3:1时会得到较好的热点分析结果。通过分析犯罪热点分布图,可协助公安机关调整警力配置,加强案件高发区警力巡逻。
[Abstract]:In order to solve the problem of optimal window width selection in crime spatial aggregation algorithm, a new clustering algorithm is proposed by combining dynamic optimization window width algorithm with DENCLUE (DENsity-based CLUstEring) algorithm. Taking the burglary case as an example, the application of this algorithm in the detection of crime hotspots is studied. The results show that the location of clustering center and the change trend of probability density can be obtained by introducing the optimized window width algorithm. When the ratio of the edge length to the distance threshold of the adjacent grid and the threshold of the distance between the adjacent points is 2:3:1, a better hot spot analysis result can be obtained. By analyzing the distribution map of crime hot spots, we can assist the public security organs to adjust the police force allocation and strengthen the police patrol in the areas with high incidence of cases.
【作者单位】： 清华大学工程物理系公共安全研究中心;
【基金】：国家自然科学基金资助项目(70773069)
【分类号】：D917
本文编号：2288140
[Abstract]:In order to solve the problem of optimal window width selection in crime spatial aggregation algorithm, a new clustering algorithm is proposed by combining dynamic optimization window width algorithm with DENCLUE (DENsity-based CLUstEring) algorithm. Taking the burglary case as an example, the application of this algorithm in the detection of crime hotspots is studied. The results show that the location of clustering center and the change trend of probability density can be obtained by introducing the optimized window width algorithm. When the ratio of the edge length to the distance threshold of the adjacent grid and the threshold of the distance between the adjacent points is 2:3:1, a better hot spot analysis result can be obtained. By analyzing the distribution map of crime hot spots, we can assist the public security organs to adjust the police force allocation and strengthen the police patrol in the areas with high incidence of cases.
【作者单位】： 清华大学工程物理系公共安全研究中心;
【基金】：国家自然科学基金资助项目(70773069)
【分类号】：D917
【相似文献】
相关期刊论文 前10条
1 颜峻;袁宏永;疏学明;钟少波;;用于犯罪空间聚集态研究的优化聚类算法[J];清华大学学报(自然科学版)网络.预览;2009年02期
2 ;[J];;年期
3 ;[J];;年期
4 ;[J];;年期
5 ;[J];;年期
6 ;[J];;年期
7 ;[J];;年期
8 ;[J];;年期
9 ;[J];;年期
10 ;[J];;年期
，本文编号：2288140
本文链接：https://www.wllwen.com/falvlunwen/fanzuizhian/2288140.html
