基于DBSCAN算法的城轨车站乘客聚集特征分析
|
摘要
发掘并掌握站内乘客群体的聚集时空变化规律,对于优化城市轨道交通线网间车辆的调度,特别是优化灾害条件下的客流组织管理等,具有积极的作用.针对具有密度分布非均匀特征的车站乘客位置数据集,提出一种基于高斯混合模型的DBSCAN聚类算法.首先,利用高斯混合模型对数据集进行密度的分层处理;然后,面向不同密度层次的数据集进行局部聚类,确定各密度层数据集的参数,并选取恰当的种子以完成局部聚类簇扩展;最后,将各密度层次数据集的聚类结果进行合并.通过标准和实测数据的计算结果表明,基于高斯混合模型优化后的DBSCAN算法,对于非均匀密度分布的乘客位置分布数据具有更好的聚类效果.
Exploring and grasping the temporal and spatial variation rules of passenger group's aggregation in the station has a positive effect on optimizing the scheduling of vehicles in the urban rail transit network, especially optimizing the organization and management of passengers under disaster conditions. In this paper, a density based spatial clustering of applications with noise(DBSCAN) clustering algorithm based on the Gaussian mixture model is proposed for the station passenger location data set with non uniform density distribution. Firstly, the Gaussian mixture model is used to process the density of data sets. Then, local clustering is performed on data sets with different density levels to determine the parameters of each density layer data set. The appropriate seeds are selected to expand the local cluster cluster. Finally, the clustering results of each density hierarchical data set are merged. Through the calculation of the standard and measured data, it is illustrated that the DBSCAN algorithm based on the Gaussian mixture model has better clustering effect for the passenger location distribution data with non-uniform density distribution.
Exploring and grasping the temporal and spatial variation rules of passenger group's aggregation in the station has a positive effect on optimizing the scheduling of vehicles in the urban rail transit network, especially optimizing the organization and management of passengers under disaster conditions. In this paper, a density based spatial clustering of applications with noise(DBSCAN) clustering algorithm based on the Gaussian mixture model is proposed for the station passenger location data set with non uniform density distribution. Firstly, the Gaussian mixture model is used to process the density of data sets. Then, local clustering is performed on data sets with different density levels to determine the parameters of each density layer data set. The appropriate seeds are selected to expand the local cluster cluster. Finally, the clustering results of each density hierarchical data set are merged. Through the calculation of the standard and measured data, it is illustrated that the DBSCAN algorithm based on the Gaussian mixture model has better clustering effect for the passenger location distribution data with non-uniform density distribution.
引文
[1]曹守华.城市轨道交通乘客交通特性分析及建模[D].北京:北京交通大学交通运输学院, 2009.(Cao S H. Analysis and mdeling on pssengers traffic characteristics for urban rail transit[D]. Beijing:School of Traffic and Transportation, Beijing Jiaotong University,2009.)
[2]宋庆梅,吴非,袁振洲.城市轨道交通站台的乘客分布特性分析及建模[J].城市轨道交通研究, 2011, 14(9):43-47.(Song Q M, Wu F, Yuan Z Z. Urban rail transit station platform passenger distribution character[J]. Urban Mass Transit, 2011, 14(9):43-47.)
[3]李灿.城市轨道交通枢纽乘客流交通特性分析及建模[D].北京:北京交通大学交通运输学院, 2008.(Li C. Study and model of the characteristic of passengers flow in urban railway transit hub[D]. Beijing:School of Traffic and Transportation, Beijing Jiaotong University,2008.)
[4]张琦,韩宝明.城市轨道交通车站乘客群体行为特征研究[J].城市交通, 2010, 8(4):41-46.(Zhang Q, Han B M. A study on characteristics of passenger collective behaviors on urban rail transit stations[J]. Urban Transport of China, 2010, 8(4):41-46.)
[5]许阳.城市轨道交通车站站台候车乘客分布规律研究[D].北京:北京交通大学交通运输学院, 2015.(Xu Y. Research on the spatial distribution of waiting passengers on the platform of a rail transit station[D].Beijing:School of Traffic and Transportation, Beijing Jiaotong University, 2015.)
[6]李思杰,朱炜,黄兆东.基于WIFI数据的城市轨道交通乘客出行时空轨迹推定[J].华东交通大学学报, 2017,34(2):85-92.(Li S J, Zhu W, Huang Z D. Travel time-space trajectory characterization of urban rail transit network based on WIFI data[J]. J of East China Jiaotong University, 2017,34(2):85-92.)
[7] Delafontaine M, Versichele M, Neutens T, et al. Analysing spatiotemporal sequences in Bluetooth tracking data[J].Applied Geography, 2012, 34(1):659-668.
[8] Yoshimura Y, Sobolevsky S, Ratti C, et al. An analysis of visitors’ behavior in the louvre museum:A study using bluetooth data[J]. Environment&Planning B Planning&Design, 2014, 41(6):1113-1131.
[9] Liebig T, Andrienko G, Andrienko N. Methods for analysis of spatio-temporal bluetooth tracking data[J]. J of Urban Technology, 2014, 21(2):81-91.
[10] Petrenko A, Sizo A, Qian W, et al. Exploring mobility indoors:An application of sensorbased and GIS systems[J]. Trans in Gis, 2014, 18(3):351-369.
[11] Xiong Zhongyang, Chen Ruotian, Zhang Yufang, et al.Multi-density DBSCAN algorithm based on density levels partitioning[J]. J of Information and Computational Science, 2011, 9(10):2739-2749.
[12]许芳芳.基于DBSCAN优化算法的Web文本聚类研究[D].上海:华东师范大学信息科学技术学院, 2011.(Xu F F. The research on web text clustering based on DBSCAN optimized algorithm[D]. Shanghai:School of Information Science Technology, East China Normal University, 2011.)
[13] Liu Zunxiong, Chen Ying, Tian Shanshan, et al.Multivariate Gaussian mixture model based clustering with truncated and censored data[J]. J of Information&Computational Science, 2015, 12(2):775-785.
[14] Nikos V, Ar Istd Is L. A greedy algorithm for Gaussian mixture learing[J]. Neural Processing Letters, 2002,15(1):77-87.
[15] Dempster A P, Laird N M, Rubin D B. Maximum likelihood from incomplete data via the EM algorithm[J]. J of the Royal Statistical Society, Series B:Methodological, 1977, 39(1):1-38.
[16]王实美.基于DBSCAN的自适应非均匀密度聚类算法研究[D].北京:北京交通大学电子信息工程学院,2017.(Wang S M. Research on adaptive varied density clustering algorithm based on DBSCAN[D]. Beijing:School of Electronic and Information Engineering,Beijing Jiaotong University, 2017.)
[17] Zhou Shui-geng, Zhou Ao-ying, Jin Wen, et al.FDBSCAN:A fast DBSCAN algorithm[J]. J of Software,2000, 11(6):735-744.
[18]许合利,牛丽君.基于层次与密度的任意形状聚类算法[J].计算机工程, 2016, 42(7):159-164.(Xu H L, Niu L J. Arbitrary shape clustering algorithm based on hierarchy and density[J]. Computer Engineering, 2016, 42(7):159-164.)
[19]谢江.针对非均匀密度环境的DBSCAN自适应聚类算法的研究[D].重庆:重庆大学计算机学院, 2015.(Xie J. A self-adaptive density-based clustering algorithm for discovering density varied clusters[D]. Chongqing:College of Computer Science, Chongqing University,2015.)
[2]宋庆梅,吴非,袁振洲.城市轨道交通站台的乘客分布特性分析及建模[J].城市轨道交通研究, 2011, 14(9):43-47.(Song Q M, Wu F, Yuan Z Z. Urban rail transit station platform passenger distribution character[J]. Urban Mass Transit, 2011, 14(9):43-47.)
[3]李灿.城市轨道交通枢纽乘客流交通特性分析及建模[D].北京:北京交通大学交通运输学院, 2008.(Li C. Study and model of the characteristic of passengers flow in urban railway transit hub[D]. Beijing:School of Traffic and Transportation, Beijing Jiaotong University,2008.)
[4]张琦,韩宝明.城市轨道交通车站乘客群体行为特征研究[J].城市交通, 2010, 8(4):41-46.(Zhang Q, Han B M. A study on characteristics of passenger collective behaviors on urban rail transit stations[J]. Urban Transport of China, 2010, 8(4):41-46.)
[5]许阳.城市轨道交通车站站台候车乘客分布规律研究[D].北京:北京交通大学交通运输学院, 2015.(Xu Y. Research on the spatial distribution of waiting passengers on the platform of a rail transit station[D].Beijing:School of Traffic and Transportation, Beijing Jiaotong University, 2015.)
[6]李思杰,朱炜,黄兆东.基于WIFI数据的城市轨道交通乘客出行时空轨迹推定[J].华东交通大学学报, 2017,34(2):85-92.(Li S J, Zhu W, Huang Z D. Travel time-space trajectory characterization of urban rail transit network based on WIFI data[J]. J of East China Jiaotong University, 2017,34(2):85-92.)
[7] Delafontaine M, Versichele M, Neutens T, et al. Analysing spatiotemporal sequences in Bluetooth tracking data[J].Applied Geography, 2012, 34(1):659-668.
[8] Yoshimura Y, Sobolevsky S, Ratti C, et al. An analysis of visitors’ behavior in the louvre museum:A study using bluetooth data[J]. Environment&Planning B Planning&Design, 2014, 41(6):1113-1131.
[9] Liebig T, Andrienko G, Andrienko N. Methods for analysis of spatio-temporal bluetooth tracking data[J]. J of Urban Technology, 2014, 21(2):81-91.
[10] Petrenko A, Sizo A, Qian W, et al. Exploring mobility indoors:An application of sensorbased and GIS systems[J]. Trans in Gis, 2014, 18(3):351-369.
[11] Xiong Zhongyang, Chen Ruotian, Zhang Yufang, et al.Multi-density DBSCAN algorithm based on density levels partitioning[J]. J of Information and Computational Science, 2011, 9(10):2739-2749.
[12]许芳芳.基于DBSCAN优化算法的Web文本聚类研究[D].上海:华东师范大学信息科学技术学院, 2011.(Xu F F. The research on web text clustering based on DBSCAN optimized algorithm[D]. Shanghai:School of Information Science Technology, East China Normal University, 2011.)
[13] Liu Zunxiong, Chen Ying, Tian Shanshan, et al.Multivariate Gaussian mixture model based clustering with truncated and censored data[J]. J of Information&Computational Science, 2015, 12(2):775-785.
[14] Nikos V, Ar Istd Is L. A greedy algorithm for Gaussian mixture learing[J]. Neural Processing Letters, 2002,15(1):77-87.
[15] Dempster A P, Laird N M, Rubin D B. Maximum likelihood from incomplete data via the EM algorithm[J]. J of the Royal Statistical Society, Series B:Methodological, 1977, 39(1):1-38.
[16]王实美.基于DBSCAN的自适应非均匀密度聚类算法研究[D].北京:北京交通大学电子信息工程学院,2017.(Wang S M. Research on adaptive varied density clustering algorithm based on DBSCAN[D]. Beijing:School of Electronic and Information Engineering,Beijing Jiaotong University, 2017.)
[17] Zhou Shui-geng, Zhou Ao-ying, Jin Wen, et al.FDBSCAN:A fast DBSCAN algorithm[J]. J of Software,2000, 11(6):735-744.
[18]许合利,牛丽君.基于层次与密度的任意形状聚类算法[J].计算机工程, 2016, 42(7):159-164.(Xu H L, Niu L J. Arbitrary shape clustering algorithm based on hierarchy and density[J]. Computer Engineering, 2016, 42(7):159-164.)
[19]谢江.针对非均匀密度环境的DBSCAN自适应聚类算法的研究[D].重庆:重庆大学计算机学院, 2015.(Xie J. A self-adaptive density-based clustering algorithm for discovering density varied clusters[D]. Chongqing:College of Computer Science, Chongqing University,2015.)