[1]刘勇,何婧,姚绍文,等.基于重心点转移的 St-DBSCAN 改进算法[J].计算机技术与发展,2018,28(11):6-11.[doi:10.3969/ j. issn.1673-629X.2018.11.002]
　LIU Yong,HE Jing,YAO Shao-wen,et al.An Improved St-DBSCAN Algorithm Based on Center of Gravity Shifting[J].,2018,28(11):6-11.[doi:10.3969/ j. issn.1673-629X.2018.11.002]
点击复制

基于重心点转移的 St-DBSCAN 改进算法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2018年11期
页码:
6-11
栏目:
智能、算法、系统工程
出版日期:
2018-11-10

文章信息/Info

Title:
An Improved St-DBSCAN Algorithm Based on Center of Gravity Shifting
文章编号:
1673-629X(2018)11-0006-06
作者:
刘勇; 何婧; 姚绍文; 向毅; 张浩
云南大学 国家示范性软件学院,云南 昆明 650500
Author(s):
LIU Yong; HE Jing; YAO Shao-wen; XIANG Yi; ZHANG Hao
National Pilot School of Software,Yunnan University,Kunming 650500,China
关键词:
时空聚类算法; St-DBSCAN 算法; 转移策略; 密度倾斜; 重心点
Keywords:
spatio-temporal clustering algorithm; St-DBSCAN algorithm; transfer strategy; density dip; center of gravity shifting
分类号:
TP302
DOI:
10.3969/ j. issn.1673-629X.2018.11.002
文献标志码:
A
摘要:
在目前已提出的聚类算法中,St-DBSCAN 算法是一种基于密度且性能优越的时空聚类算法。但是当时空点分布出现密度倾斜时,St-DBSCAN 算法会出现聚类时间过长和聚类效果不好的问题。基于此,通过对空间点分布存在的三种数据倾斜,采用数据重心点转移策略,提出了对应的解决方案,以此实现了改进后的 St-DBSCAN 算法。为了验证改进后算法的性能,以昆明市出租车 GPS 数据为实验数据,进行了算法性能对比实验。 实验结果表明,改进 St-DBSCAN 算法的时间性能和聚类效果有了一定程度的提升。
Abstract:
In the presented clustering algorithms,the St-DBSCAN is a spatio-temporal clustering algorithm based on density with superior performance. However,when the spatial distribution is tilted,the St-DBSCAN algorithm may produce too long clustering time and poor clustering effect. Based on the problem,we propose the corresponding solution by using the data center point transfer strategy for three kinds of data skew in spatial point distribution,and then implement the improved St-DBSCAN algorithm. In order to verify the proposed algorithm,the GPS data of taxi of Kunming is used as experimental data for performance comparison,which shows that the improved St-DBSCAN algorithm is improved in time performance and clustering effect.
更新日期/Last Update: 2018-11-10