摘    要：入室盗窃犯罪不仅会造成财产损失，还易诱发强奸、抢劫、杀人等暴力犯罪，是影响社会治安的重要问题之一。本文选取杭州市八城区作为研究区域，以2014年1月至2015年6月的入室盗窃案件（5203起）作为研究对象，在总结国内外犯罪时空分析方法及研究现状的基础上，分析了杭州市八区的入室盗窃案件的时空分布模式。首先，对入室盗窃犯罪的犯罪率、犯罪密度进行分析。其次，从多个层次进行时间分析：以周内变化分析，发现周一至周五犯罪案件数趋于平稳，周六周日则有所下降；以月份变化分析，发现2月份犯罪案件数最低，5月份犯罪案件数最高；另外，分析了入室盗窃案件与法定节假日的相关性，发现法定节假日的日均犯罪数明显少于非法定节假日日均犯罪数。最后，利用核密度估计法、全局空间自相关和局部空间自相关法对入室盗窃案件进行空间分布特征分析，发现全区核密度热点主要分布在下城区石桥街道、拱墅区大关街道、西湖区翠苑街道以及萧山区城厢街道；两种全局空间自相关分析方法得出的聚类特征略有差异，利用局部空间自相关分析进一步分析杭州市内部入室盗窃案件的聚集特征，发现入室盗窃案件高发聚集区主要位于三墩镇、笕桥镇、东新街道等区域，义蓬镇、党湾镇、黄湖镇等区域为案件低发聚集区。

Abstract：Burglary crime not only causes property losses, but also induces violent crimes such as rape, robbery, murder and so on. It is one of the important problems that affect public security. Based on the burglary crime data from January 2014 to June 2015, the crime rate, crime density as well as spatial-temporal distribution of burglary crime in eight districts of Hangzhou were analyzed. Firstly, we analyzed the crime rate and density of burglary crime. Then, time analysis was carried out from multiple levels. Time analysis was carried out from multiple levels; the smallest number of criminal cases was in February, while the largest number of criminal cases happened in May; in addition, the correlation between burglaries and statutory holidays was analyzed, and the daily average number of crimes in holidays is significantly less than that of workdays. Finally, the spatial distribution characteristics of burglary cases were analyzed by means of kernel density analysis, global spatial autocorrelation and local spatial autocorrelation. We found that the kernel density hot spots were mainly distributed in Shiqiao community in Xiacheng district, Daguan community in Gongshu district, Cuiyuan community in Xihu district and Chengxiang community in Xiaoshan district. The clustering characteristics of the two global spatial autocorrelation analysis methods were slightly different. We used the local spatial autocorrelation analysis to analyze the characteristics of the burglaries in Hangzhou, it was found that the high incidence area of burglaries was mainly located in San Dun towns, Jian Qiao Town, Dongxin Street and so on, and Yi Peng Town, Dang Wan Town, Huanghu town and other towns were low incidence areas.

关键词： 地理信息系统；入室盗窃；时空分布；核密度估计；空间自相关

Keywords：GIS; Burglary crime; Spatio-temporal distribution; Kernel density estimation；Spatial autocorrelation

目录

1. 引言 1

1.1 研究背景及意义 1

1.2 国内外研究现状 2

1.3 本文研究技术路线 3

1.4 论文组织结构 4

2. 研究区域和数据 5

2.1 研究区与数据源 5

2.2 数据预处理 6

3. 研究方法 7

3.1 核密度估计法 7

3.2 空间自相关方法 7

4. 犯罪时空分布模式分析 8

4.1 犯罪率与犯罪密度分析 8

4.2 入室盗窃案件时间分布分析 9

4.2.1 按月统计分析 10

4.2.2 按周统计分析 10

4.2.3 按法定节假日分析 11