National small area classifications in Britain were first produced over 40 years ago using statistics from 1971 Population Census and have now become a regular feature of governmental, academic and commercial analysis of census information. These classifications aim to encapsulate the aggregate demographic and socio-economic character of small areas by means of a simple thumbnail description. However, these pen portraits often also refer to the environmental nature of the different types of area where people live, employing terms such as ‘leafy suburb’, ‘industrial hinterland’ or ‘agricultural heartland’. This paper reports on research that aims to determine whether a set of environmental (land use) indicators are capable of discriminating between areas in a way that matches a ‘standard’ area classification derived from multivariate analysis of demographic and socio-economic statistics. The research assesses the impact of adding a set of environmental (land use) variables to a collection of Census variables on area classification using k-means clustering varia.in two contrasting case study local authorities. The results reveal that clustering with and without the addition of land use variables produce partially overlapping (coincident) classifications of the small areas and certain of the land use variables are aligned with some area types.

中文翻译：

---

住宅区分类：一种环境改善的方法

英国的全国小区域分类最早是在 40 多年前使用 1971 年人口普查的统计数据产生的，现在已成为政府、学术和商业分析人口普查信息的常规特征。这些分类旨在通过简单的缩略图描述来概括小区域的总体人口统计和社会经济特征。然而，这些钢笔画也常常指人们居住的不同类型地区的环境性质，使用“绿树成荫的郊区”、“工业腹地”或“农业中心地带”等术语。本文报告的研究旨在确定一组环境（土地利用）指标是否能够以与从人口统计和社会经济统计的多变量分析得出的“标准”区域分类相匹配的方式区分区域。该研究使用 k 均值聚类变量在两个对比案例研究地方当局评估了将一组环境（土地利用）变量添加到人口普查变量集合对区域分类的影响。结果表明，添加和不添加土地利用变量的聚类会产生小区域的部分重叠（重合）分类，并且某些土地利用变量与某些区域类型对齐。该研究使用 k 均值聚类变量在两个对比案例研究地方当局评估了将一组环境（土地利用）变量添加到人口普查变量集合对区域分类的影响。结果表明，添加和不添加土地利用变量的聚类会产生小区域的部分重叠（重合）分类，并且某些土地利用变量与某些区域类型对齐。该研究使用 k 均值聚类变量在两个对比案例研究地方当局评估了将一组环境（土地利用）变量添加到人口普查变量集合对区域分类的影响。结果表明，添加和不添加土地利用变量的聚类会产生小区域的部分重叠（重合）分类，并且某些土地利用变量与某些区域类型对齐。