基于PAM聚类的轨道交通站点画像分析
Portrait Analysis of Rail Transit Stations Based on PAM Clustering
洪英 1李喆康 2李旭2
作者信息
- 1. 苏州市轨道交通集团有限公司
- 2. 南京市城市与交通规划设计研究院
- 折叠
摘要
文章基于多源数据融合方法,从客流水平、居民需求、接驳特性、周边设施4个维度对轨道站点特征进行提取与构建.采用PAM算法对站点进行聚类研究,最终形成6类站点画像的标签.在此基础上探索站点功能定位、客流模式之间的内在关系,梳理不同类型站点在城市化进程中面临的阶段性矛盾,指导客流预测任务、引流策略制定等后续工作.
Abstract
Based on multi-source data fusion,this paper extracts and constructs the characteristics of rail stations from four dimensions:passenger flow level,resident demand,connection characteristics,and surrounding facilities.The PAM algorithm was used to study the clustering of the sites,and finally the labels of six types of site portraits were formed.On this basis,the internal relationship between the functional positioning of stations and passenger flow patterns is explored,and the phased contradictions faced by different types of stations in the process of urbanization are sorted out,so as to provide guidance for follow-up research such as passenger flow prediction and drainage measures.
关键词
多源数据/轨道站点画像/PAM聚类Key words
multi-source data/portrait analysis of rail transit stations/PAM clustering引用本文复制引用
出版年
2024
NETL
NSTL
万方数据