陈思宇 孟斌 李梦天

[摘要]空间同位模式是指在一定空间范围内频繁发生并且位置相邻的事件集合，对于研究不同空间要素之间交互关系具有重要意义。城市中职住要素的研究是很典型的同位模式。基于Ripley′s K函数基础上变形的双变量K（r）函数和双变量g（r）函数，对获取到的北京市五环内职住要素和交通要素进行空间同位模式分析，最终得出：1）住宅区和写字楼之间的同位模式较为复杂，低价位住宅区与写字楼之间不存在同位模式;中等价位住宅区和高价

位住宅区与写字楼存在同位模式，但尺度范围不同;2）不同价位住

宅区与地铁站之间都在0～2 km范圍上具有空间同位模式，并且随着距离的增加，住宅区与地铁站的空间同位模式强度在不断减弱;3）写字楼与地铁站点同样在中距离范围内具有空间同位模式，并且随着距离的增加，空间同位模式的强度均在减小。

[关键词]空间同位模式;Ripley′s K函数;职住关系

[中图分类号]TU 984[文献标志码]A[文章编号]10050310（2021）01002009

Research on the Elements of Beijing′s Urban JobsHousing

Based on the Colocation Model

Chen Siyu， Meng Bin，Li Mengtian

（College of Applied Arts and Science， Beijing Union University， Beijing 100191， China）

Abstract： Spatial colocation mode refers to a collection of events that frequently occur within a certain spatial range and are located adjacent to each other， which is of great significance to the study of the interaction between different spatial elements. The research on the elements of the urban jobshousing is a typical colocation model. Based on the deformed bivariate K（r） function and bivariate g（r） stemming from Ripley′s K function， this paper analyzes the spatial colocation pattern of the acquired elements of jobshousing and traffic within the Fifth Ring Road of Beijing， and finally obtains： 1） The colocation pattern between residential buildings and office buildings is complicated. There is no colocation pattern between lowpriced residential areas and office buildings; mediumpriced residential areas and highpriced residential areas have colocation patterns relationship with office buildings， but the scale ranges are different. 2） There is a spatial colocation pattern between residential buildings of different prices and subway stations in the range of 0～2 km. As the distance increases， the intensity of the spatial colocation pattern between residential buildings and subway stations constantly weakens. 3） Office buildings and subway stations also have spatial colocation patterns in the middledistance range. As the distance increases， the intensity of the spatial colocation patterns decreases.

Keywords： Colocation pattern; Ripley′s Kfunction; Relationship of jobshousing

引言

空间同位模式是指频繁地出现在同一区域的一组空间要素或事件[12]。例如，生态学中的尼罗鳄与埃及燕鸻、经济学中的供应商与制造商的分布，均是典型的同位模式[3]。空间同位模式的挖掘是空间关联性挖掘的一种，逐渐成为空间挖掘中的一项重要任务[45]，有助于揭示地理现象或事件的分布规律。空间同位模式挖掘的方法主要可分为空间统计方法和数据挖掘方法两类[6]，其中，空间统计方法是通过关联性指标去描述不同类空间要素之间的关系，这些指标包括交叉K函数、平均最近邻距离、空间回归模型、同位系数、Q 统计等[712]。……