支持城市功能街区划分的有序语义聚类算法  

An Ordered Semantic Clustering Algorithm Supporting Urban Block Knowledge Graph

作　　者：张冉冉 刘俊岭[1,2] 孙焕良 许景科[1,2] 

机构地区：[1]沈阳建筑大学计算机科学与工程学院,辽宁 沈阳 [2]辽宁省城市建设大数据管理与分析重点实验室,辽宁 沈阳

出　　处：《数据挖掘》2024年第1期10-19,共10页Hans Journal of Data Mining

摘　　要：城市中的功能大都分布在沿街道的两侧建筑,表现为线性街区,识别城市街区功能划分的特征可为城市空间结构及资源的全面规划、合理配置、统筹安排等提供帮助。传统线性语义聚类算法可用于划分单功能城市街道区,但城市街区不仅包括单一功能分区,还包括混合功能区。本文提出一种支持城市功能街区划分的有序语义聚类算法,在发现单一功能区的同时,也发现混合区并定义了一种新的度量混合功能区的方法。提出的算法基于层次聚类思想,具体算法分为两阶段,第一阶段为层次树生成,采用凝聚的方法将相邻的相似分段合并,得到层次树;第二阶段为功能区提取,进行单一功能区与混合功能区识别,获取给定街区的线性功能区。在真实数据集上的实验结果表明,所提出的算法可以有效发现混合功能区。The functions in a city are mostly distributed along the buildings on both sides of the street, manifested as linear blocks. Identifying the characteristics of the functional division of urban blocks can provide assistance for the comprehensive planning, rational allocation, and overall arrangement of urban spatial structure and resources. The traditional linear semantic clustering algorithm can be used to divide the single function urban street area, but the city block includes not only the single function area, but also the mixed function areas. This article proposes an ordered semantic clustering algorithm that supports the division of urban functional blocks. While discovering a single functional area, it also discovers mixed areas and defines a new method for measuring mixed functional areas. The proposed algorithm is based on the idea of hierarchical clustering, which is divided into two stages. The first stage is the generation of a hierarchical tree, which uses the aggregation method to merge adjacent similar segments to obtain a hierarchical tree. The second stage involves extracting functional areas, identifying single and mixed functional areas, and obtaining linear functional areas for a given block. The experimental results on real datasets show that the proposed algorithm can effectively discover mixed functional areas.

关 键 词：POI 有序聚类 功能区划分 混合功能区 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]