提出了一种基于连通树的热区检测算法, 可检测任意形状的热区, 目的是通过检测兵棋推演过程中军事行动频繁的区域, 为受训人员了解整个战场态势提供辅助参考. 算法在明确了热区定义的基础上, 首先构建连通树将数据集按连通区域进行最小划分, 再根据设定的密度阈值对连通树进行剪枝. 剪枝处理后的每棵连通子树是最终的热区. 理论和实验结果均验证了该算法的有效性.
Abstract
A hotspot detection algorithm based on connected tree is proposed, which is capable of detecting arbitrarily shaped hotspots during the wargaming process. By detecting the areas with high concentrations of martial events, this algorithm could assist trainees understanding the whole wargaming battlefield situation. After making the definition of a hotspot, a connected tree is built in order to least divide the whole dataset into connected regions, and a pruning procedure is carried out according to the provided density threshold value. Each pruned connected subtree is a hotspot which we would like to acquire. Both the theoretical analysis and experimental results verify the effectiveness of the algorithm.
关键词
兵棋推演 /
连通区域 /
热区检测 /
连通树
{{custom_keyword}} /
Key words
wargaming /
connected area /
hotspot detection /
connected tree
{{custom_keyword}} /
中图分类号:
O22
{{custom_clc.code}}
({{custom_clc.text}})
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 胡晓峰, 杨镜宇, 司光亚, 等.战争复杂系统仿真分析与实验[M].北京:国防大学出版社, 2008. Hu X F, Yang J Y, Si G Y, et al. War Complex System Simulation Analysis & Experimentation[M]. Beijing: Press of National Defence University, 2008.
[2] Openshaw S, Charlton M, Wymer C, et al. A mark 1 geographical analysis machine for the automated analysis of point data sets[J]. International Journal of Geographical Information Science, 1987, 1(4): 335–358.
[3] Block R. Gang activity and overall levels of crime: A new mapping tool for defining areas of gang activity using police records[J]. Journal of Quantitative Criminology, 2000, 16(3): 369–383.
[4] Kulldorff M, Huang L, Pickle L, et al. An elliptic spatial scan statistic[J]. Statistics in Medicine, 2006, 25: 3929–3943.
[5] Patil G, Taillie C. Upper level set scan statistic for detecting arbitrarily shaped hotspots[J]. Environmental and Ecological Statistics, 2004, 11: 183–197.
[6] Modarres R, Patil G P. Hotspot detection with bivariate data[J]. Journal of Statistical Planning and Inference, 2007, 137: 3643–3654.
[7] Getis A. Reflections on spatial autocorrelation[J]. Regional Science and Urban Economics, 2007, 37(4): 491–496.
[8] Anselin L. Local indicators of spatial association — LISA[J]. Geographical Analysis, 1995, 27: 93–115.
[9] Keith O J, Getis A. Local spatial autocorrelation statistics: Distributional issues and an application[J]. Geographical Analysis, 1995, 27(4): 286–306.
[10] Benoit F, Mouchart M, Martin E S, et al. The local spatial autocorrelation and the kernel method for identifying black zones: A comparative approach[J]. Accident Analysis and Prevention, 2003, 35: 991–1004.
[11] Steenberghen T, Dufays T, Thomas I, et al. Intra-urban location and clustering of road accidents using GIS: A Belgian example[J]. International Journal of Geographical Information Science, 2004, 18: 169–181.
[12] Massimo C, Haining R, Wiles P. A comparative evaluation of approaches to urban crime pattern analysis[J]. Urban Studies, 2000, 37: 711–729.
[13] Lawson A B. Hotspot detection and clustering: Ways and means[J]. Environmental and Ecological Statistics, 2010, 17: 231–245.
[14] Grubesic T H. On the application of fuzzy clustering for crime hot spot detection[J]. Journal of Quantitative Criminology, 2006, 22: 77–105.
[15] Duque J C, Ramos R, Surinach J, et al. Supervised regionalization methods: A survey[J]. International Regional Science Review, 2007, 30: 195–220.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国防预研基金(9140A04040109KG);中国博士后科学基金(201003746)
{{custom_fund}}