Abstract: Using spatial analysis methods, the geographic pattern of pickpocket incidents along the Chang'an Street in Beijing was examined in this study. First, the crime distribution along the street was identified. The results demonstrate that major crime clustering areas existed in the Xidan business area (A), Jianguomen area (B), and Dawanglu-Sihui Area (C). By comparing the spatial pattern of crimes with population density and point of interest (POI) density along the street, it was found that crimes tended to be clustered around locations having higher POI density. In the next step, spatiotemporal patterns of offences in the three areas were analyzed using kernel density estimation and space-time hotspot matrix. The results indicate that zone A maintained higher crime level between 10 a.m. and 6 p.m. and the peak time appeared at 12 a.m., and the offences concentrated in the major shopping malls. However, in zones B and C, the higher level of crimes occurred in the hours around 6 a.m. and 6 p.m., which are the periods of peak traffic flow in the morning and evening. Lastly, some detailed crime prevention and suppression suggestions and strategies are proposed on the basis of the spatial attributes and blind areas theory.

Key words: pickpocket, crime hotspot, spatial analysis, spatiotemporal hotspot matrix, Beijing

摘要： 本文利用空间分析方法,对北京市长安街沿线的扒窃类案件的空间分布特征进行了研究。首先分析了长安街沿线的犯罪分布状况,发现该长安街沿线共存在3个异常明显的犯罪高发区域,分别对应于西单商业圈(A)、建国门路段(B)和大望路—四惠路段(C),通过将长安街沿线的常住人口与兴趣点(Point of Interest, POI)分布进行比较,发现案件与POI密度较高的区域存在较高的相关性。随后,利用核密度估计方法和时空热点矩阵方法分析了3个异常明显的案件高发区域内的犯罪活动时空分布模式,结果表明：高发区域A的犯罪高发时段为上午10时至下午18时,其中峰值出现在正午12时,犯罪活动主要集中在区域内的一些大型购物场所内;高发区域B和C的犯罪高发时间均为上午6时和下午18时前后,与早晚交通高峰时间比较一致,犯罪活动主要集中在区域内比较重要的公共交通站点附近。最后,根据长安街沿线案件高发区域内犯罪活动的时空规律性特征,提出了犯罪防控措施与建议。

关键词: 扒窃, 犯罪热点, 空间分析, 时空热点矩阵, 北京