摘要
时空大数据是目前研究的热点。如何从海量手机信令数据中获取有价值的信息是研究手机信令数据的难点。本文在基于距离的点聚合方法的基础上,提出了将基于密度聚类算法DBSCAN与基于距离聚类算法kmeans相结合的点聚合算法。采用DBSCAN与kmeans相结合的点聚合算法实现手机信令数据的可视化,不仅能避免手机信令数据在可视化时点数据的堆叠和覆盖问题,而且使得其聚合后获取数据的空间分布结构更准确。
Spatio-temporal big data is a hot research topic nowadays.How to obtain valuable information from massive signaling data of mobile phone is a difficult problem in the research of mobile phone signaling data.Based on the distance-based point aggregation method,this paper proposes a point aggregation algorithm which combines the densitybased clustering algorithm DBSCAN with the distance-based clustering algorithm kmeans.The point aggregation algorithm based on DBSCAN and kmeans is used to realize the visualization of mobile signaling data.It not only avoids the problem of stacking and overlaying the point data in visualization,but also makes the spatial distribution structure of the data acquired after aggregation more accurate.
作者
史新颖
夏元平
毛曦
殷红梅
SHI Xinying;XIA Yuanping;MAO Xi;YIN Hongmei(Faculty of Geomatics,East China University of Technology,Nanchang Jiangxi,330013,China;Chinese Academy of Surveying and Mapping,Beijing 100039,China)
出处
《北京测绘》
2019年第2期132-137,共6页
Beijing Surveying and Mapping
基金
中国测绘学院研究院基本科研业务经费项目(7771802
7771721)
江西省星火计划项目(20161BBB29002)
关键词
手机信令数据
聚类
点位误差
cell phone signaling data
clustering
point error
