北京中心城区绅士化空间定量识别与类型特征关联数据

绅士化是具有较高社会经济地位群体在一定空间内的聚集现象和社会空间高端化重构过程，国内既有研究更多关注于绅士化特征与过程描述，以及由此延展的机制与效应分析，较少以计量手段展开城市范围精细尺度的绅士化空间识别与类型划分研究。本文以北京五环路围合的中心城区为例，基于百度地图慧眼时空大数据（百度慧眼）提供的100 m网格居民社会经济属性数据库，选取收入水平、教育层次和职业特征等相关指标，采用半监督聚类和集成学习方法，量化识别具有典型绅士化特征的城市空间，并利用遗传增长算法开展绅士化类型分析并比较类型间差异。

Gentrification refers to the phenomenon of agglomeration of groups with higher socioeconomic status within a specific spatial scope and the process of high-end social space restructuring. Existing domestic studies mostly focus on describing the characteristics and processes of gentrification, as well as the ensuing analysis of its mechanisms and effects, while few have carried out research on spatially identifying and classifying the typologies of urban gentrification at fine-grained scales across the entire urban area using quantitative methods. This paper takes the central urban area enclosed by the 5th Ring Road of Beijing as the study case. Based on the 100-meter grid resident socioeconomic attribute database provided by Baidu Maps Wisdom Spatiotemporal Big Data (Baidu Wisdom), this study selects relevant indicators including income level, educational attainment and occupational characteristics, adopts semi-supervised clustering and ensemble learning methods to quantitatively identify urban spaces with typical gentrification characteristics, and utilizes the genetic growth algorithm to conduct gentrification typology analysis and compare the differences among various types.

搜集汇总

背景与挑战

背景概述

该数据集聚焦于2016-2023年北京五环路内的中心城区，基于百度慧眼提供的100米网格社会经济属性大数据，采用半监督聚类和集成学习等计量方法，定量识别了具有绅士化特征的城市空间并进行了类型划分。数据集旨在弥补国内相关研究在精细尺度定量分析方面的不足，为城市地理学领域的绅士化现象研究提供了高分辨率、跨期可比的空间分析数据。