邸少宁,朱杰,郑加柱,杨静,丁凯孟

• 1:南京林业大学土木工程学院
• 2:南京师范大学地理科学学院
• 3:金陵科技学院网络与通信工程学院

摘要(Abstract)：

针对现有出租车轨迹数据挖掘中时间序列邻近度量方法存在的问题,提出一种基于DBSCAN算法和改进的DTW距离的时间序列聚类算法提取具有相似性出行特征的时空模式,进而研究城市人群出行行为的时空差异。以南京市为例,结合电子地图对出行模式的空间分布特征进行分析,证明了本文所提出的方法的有效性。实验结果表明:在空间分布上,工作日出租车出行模式按照平均出行频次由高到低排序,从城市中心向四周扩散,呈中心环状分布,出行模式区域界限较为明显,同类出行模式分布区域对应相似的功能。提出了一种基于DBSCAN算法和改进的DTW距离的时间序列聚类算法提取具有相似性出行特征的时空模式,有效地分析城市人群出行行为的时空差异。

关键词(KeyWords)： 出租车轨迹数据;时间序列度量;时间序列聚类;出行模式

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学青年科学基金项目（41501431）;; 江苏省自然科学青年基金项目（BK200170116）;; 南京林业大学人才科研启动基金项目（GXL2018049）

作者(Author): 邸少宁,朱杰,郑加柱,杨静,丁凯孟

DOI: 10.16251/j.cnki.1009-2307.2021.01.028

参考文献(References)：

• [1] 杨喜平,方志祥,尹凌.城市空间结构要素与人群聚散稳定性的关联性探索[J].地球信息科学学报,2018,20(6):791-798.(YANG Xiping,FANG Zhixiang,YIN Ling.Exploring the relationship between urban spatial structure and the stability of human convergence[J].Journal of Geo-information Science,2018,20(6):791-798.)
• [2] YU Wenhao,GUAN Menglin,CHEN Zhanlong.Analyzing spatial community pattern of network traffic flow and its variations across time based on taxi GPS trajectories[J].Applied Sciences,2019,9(10):2054.
• [3] ZHANG Dongzhi,LEE Kyungmi,LEE Ickjai.Semantic periodic pattern mining from spatio-temporal trajectories[J].Information Sciences,2019,502:164-189.
• [4] 李佳蓉,夏昊,张迎,等.城市居民出行O/D时空分布特征的轨迹数据提取[J].测绘科学,2020,45(2):150-158.(LI Jiarong,XIA Hao,ZHANG Ying,et al.Trajectory data extract of O/D temporal and spatial distribution of urban residents[J].Science of Surveying and Mapping,2020,45(2):150-158.)
• [5] 陈玲燕.出租车GPS数据的信息挖掘研究[D].北京:北京交通大学,2018.(CHEN Lingyan.Research on information mining of taxi GPS data[D].Beijing:Beijing Jiaotong University,2018.)
• [6] 张明月.基于出租车轨迹的载客点与热点区域推荐[D].长沙:湖南科技大学,2013.(ZAHNG Mingyue.Recommendation for pick-up points and hot zone based on taxi trajectories[D].Changsha:Hunan University of Science and Technology,2013.)
• [7] 蔡柳.基于出租车轨迹数据挖掘的居民出行特征研究[D].西安:长安大学,2017.(CAI Liu.Resident trip feature research based on taxi trajectory data mining[D].Xi’an:Chang’an Univeristy,2017.)
• [8] 马云飞.基于出租车轨迹点的居民出行热点区域与时空特征研究[D].南京:南京师范大学,2014.(MA Yunfei.Research on residents’ behavior of attractive areas and spatio-temporal feature based on taxi trajectory data:a case of Kunshan city[D].Nanjing:Nanjing Normal University,2014.)
• [9] 岳梦雪.多源轨迹数据时空聚类及居民出行模式分析[D].武汉:武汉大学,2017.(YUE Mengxue.Spatio-temporal clustering and resident trip mode analysis on multi-source trajectory data[D].Wuhan:Wuhan University,2017.)
• [10] JEUNG H,YIU M L,JENSEN C S.Trajectory pattern mining[G]//Computing with Spatial Trajectories,[S.l.:s.n.],2011.
• [11] SHEN Ying,ZHAO Ligang,FAN Jing.Analysis and visualization for hot spot based route recommendation using short-dated taxi GPS traces[J].Information,2015,6(2):134-151.
• [12] 程智源.基于轨迹聚类的交通热点分析[D].成都:电子科技大学,2018.(CHENG Zhiyuan.Traffic hotspot analysis based on trajectory clustering[D].Chengdu:University of Electronic Science and Technology of China,2018.)
• [13] ZHANG Daqiang,WAN Jiafu,HE Zongjian,et al.Identifying region-wide functions using urban taxicab trajectories[J].ACM Transactions on Embedded Computing Systems,2016,15(2):36.
• [14] 姜海林.基于出租车数据的城市居民出行时空特征研究[D].武汉:武汉大学,2015.(JIANG Hailin.Study on the temporal and spatial characteristics of urban residents’travel based on taxi data[D].Wuhan:Wuhan University,2015.)
• [15] 毕硕本,万蕾,杨树亮,等.基于GPS数据的南京出租车上下客时间特征及热点时空分布[J].中国科技论文,2018,13(9):1023-1028.(BI Shuoben,WAN Lei,YANG Shuliang,et al.Spatial-temporal distribution pf the pick-up and drop-off of taxi passengers in Nanjing based on GPS data[J].China Science Aper,2018,13(9):1023-1028.
• [16] 谷岩岩,焦利民,董婷,等.基于多源数据的城市功能区识别及相互作用分析[J].武汉大学学报(信息科学版),2018,43(7):1113-1121.(GU Yanyan,JIAO Limin,DONG Ting,et al.Spatial distribution and interaction analysis of urban functional areas based on multi-source data[J].Wuhan:Geomatics and Information Science of Wuhan University,2018,43(7):1113-1121.)
• [17] HU Yujie,MILLER H J,LI Xiang.Detecting and analyzing mobility hotspots using surface networks[J].Transactions in GIS,2014,18(6):911-935.
• [18] 吴华意,黄蕊,游兰,等.出租车轨迹数据挖掘进展[J].测绘学报,2019,48(11):1341-1356.(WU Huayi,HUANG Rui,YOU Lan,et al.Recent progress in taxi trajectory data mining[J].Acta Geodaetic et Cartographica Sinica,2019,38(11):1341-1356.)
• [19] 腾巧爽,孙尚宇,秘金钟.众源地理空间数据的城市热点区域探测[J].测绘科学,2018,43(5):74-80.(TENG Qiaoshuang,SUN Shangyu,BEI Jingzhong.Urbanhot pots detection based on crowdsourcing geospatial data[J].Science of Surveying and Mapping,2018,43(5):74-80.)
• [20] 陈婷.基于移动时空轨迹的路网热点区域挖掘系统设计与实现[D].成都:电子科技大学,2019.(CHEN Ting.Design and implementation of hotspot area mining system for road network based on mobile spatio-temporal trajectory[D].Chengdu:University of Electronic Science and Technology of China,2019.)
• [21] SHEN,LIU,CHEN.Discovering spatial and temporal patterns from taxi-based floating car data:a case study from Nanjing[J].GIScience &Remote Sensing,2017,54(5):617-638.
• [22] 伏家云,靖常峰,杜明义.空间密度聚类模式挖掘方法DBSCAN研究回顾与进展[J].测绘科学,2018,43(12):50-57.(FU Jiayun,JING Changfeng,DU Mingyi.Review and progress of DBSCAN research on spatial density clustering pattern mining method[J].Science of Surveying and Mapping,2018,43(12):50-57.)
• [23] XI Liu,LI Gong,YONGXI Gong,YU Liu.Revealing travel patterns and city structure with taxi trip data[J].Journal of Transport Geography,2015,43:78-90.
• [24] 陈海燕,刘晨晖,孙博.时间序列数据挖掘的相似性度量综述[J].控制与决策,2017,32(1):1-11.(SONG Haiyan,LIU Chenhui,SUN Bo.Survey on similarity measurement of time series data mining[J].Control and Decision,2017,32(1):1-11.)
• [25] YANG Jing,SUN Yizhong,SHANG Bowen,et al.Understanding collective human mobility spatiotemporal patterns on weekdays from taxi origin-destination point data[J].Sensors (Basel,Switzerland),2019,19(12):2812.
• [26] 李海林,郭崇慧.时间序列数据挖掘中特征表示与相似性度量研究综述[J].计算机应用研究,2013,30(5):1285-129.(LI Hailin,GUO Chonghui.Survey of feature representations and similarity measurements in time series data mining[J].Application Research of Computers,2013,30(5):1285-129.)
• [27] HENNIG C,HAUSDORF B.Design of dissimilarity measures:a new dissimilarity between species distribution areas[G]//Data Science and Classification.[S.l.:s.n.],2006.
• [28] CHOUAKRIA A D,NAGABHUSHAN P N.Adaptive dissimilarity index for measuring time series proximity[J].Advances in Data Analysis & Classification,2007,1(1):5-21.
• [29] MONTERO P,VILAR J A.Tsclust:an R package for time series clustering [J].Journal of Statistical Software,2014,62,1-43.
• [30] ROUSSEEUW P J.Silhouettes:A graphical aid to the interpretation and validation of cluster analysis[J].Journal of Computational and Applied Mathematics,1987.
• [31] 王宇.基于公共交通大数据的上海市居民出行时空特征研究[D].济南:山东师范大学,2018.(WANG Yu.Study on the temporal and spatial characteristics of shanghai residents’ travel based on public transport big data[D].Jinan:Shandong Normal University,2018.)