基于改进谱聚类的城市热点区域挖掘  

Urban Hotspot Area Mining Based on Improved Spectral Clustering

作　　者：王田雨 吉立新 李邵梅[2] 张建朋[2] WANG Tianyu;JI Lixin;LI Shaomei;ZHANG Jianpeng(School of Cyberspace Security,Zhengzhou University,Zhengzhou,450001,China;Information Engineering University,Zhengzhou 450001,China)

机构地区：[1]郑州大学网络空间安全学院,河南郑州450001 [2]信息工程大学,河南郑州450001

出　　处：《信息工程大学学报》2022年第3期313-319,共7页Journal of Information Engineering University

基　　金：国家自然科学基金青年科学基金资助项目(62002384)。

摘　　要：随着信息技术的快速发展,能够获取人们大量的轨迹数据。通过轨迹数据挖掘出城市热点区域,对城市规划、交通管理和一些基于位置的服务都具有重要意义。针对已有的数据挖掘方法,如k均值聚类算法、基于密度的聚类(Density Based Spatia Clustering of Applications with Noise,DBSCAN)算法、谱聚类(Spectral Clustering,SC)算法、密度峰值聚类(Density Peak Clustering,DPC)算法等,存在非凸数据集上聚类效果较差,对初始中心敏感,参数选择困难,时间复杂度较高等问题,提出基于网格划分和DPC改进谱聚类算法用于城市热点区域分析。在合成数据集上的实验结果表明,其聚类精度和时间与经典聚类算法相比具有一定的提升。在西安市出租车数据集上的实验结果表明,提出的算法能有效挖掘出城市热点区域。With the rapid development of information technology,a large amount of people’s trajectory data can be obtained.Mining urban hot spots through trajectory data is of great significance to urban planning,traffic management,and some location-based services.For existing data mining methods such as k-means、DBSCAN、SC and DPC algorithm,the clustering effect on non-convex data sets is poor,e.g.,being sensitive to the initial center,being difficult to select parameters,and relatively high time complexity.This paper proposes an improved spectral clustering algorithm based on grid division and DPC for the analysis of urban hot spots.Experimental results on the synthetic data set show that its clustering accuracy and time have a certain improvement compared with classic clustering algorithms.Experimental results on the taxi data set of Xi’an show that the algorithm proposed in this paper can effectively dig out the urban hot spots.

关 键 词：轨迹数据 数据挖掘 聚类 城市热点区域 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]