1.中山大学智能工程学院 / 广东省智能交通系统重点实验室,广东 深圳 518107
2.佳都科技集团股份有限公司,广东 广州 510665
黄敏(1975年生),女;研究方向:可计算路网;E-mail:Huangm7@mail.sysu.edu.cn
收稿:2025-06-05,
修回:2025-07-05,
录用:2025-07-05,
网络出版:2025-09-23,
纸质出版:2025-11-25
移动端阅览
黄敏,梁宁晨,王晓聪等.耦合Word2Vec和动态语义地图的车辆轨迹相似性度量[J].中山大学学报(自然科学版)(中英文),2025,64(06):76-85.
HUANG Min,LIANG Ningchen,WANG Xiaocong,et al.Vehicle trajectory similarity measures for coupling Word2Vec and dynamic semantic map[J].Acta Scientiarum Naturalium Universitatis Sunyatseni,2025,64(06):76-85.
黄敏,梁宁晨,王晓聪等.耦合Word2Vec和动态语义地图的车辆轨迹相似性度量[J].中山大学学报(自然科学版)(中英文),2025,64(06):76-85. DOI: 10.13471/j.cnki.acta.snus.ZR20250097.
HUANG Min,LIANG Ningchen,WANG Xiaocong,et al.Vehicle trajectory similarity measures for coupling Word2Vec and dynamic semantic map[J].Acta Scientiarum Naturalium Universitatis Sunyatseni,2025,64(06):76-85. DOI: 10.13471/j.cnki.acta.snus.ZR20250097.
提出一种耦合改进Word2Vec方法和动态语义地图的轨迹相似性度量方法。借助增加目的地约束的Word2Vec模型学习卡口序列关联关系,并与目的地建立显式联系;同时,动态语义地图可以作为时间和出行行为维度的相似性度量方法构建基础。实验结果表明,城市功能区在一天之中呈现出显著的动态变化特征。并且,在轨迹层次聚类任务中,本文方法的平均AC值较对比方法降低了0.36,体现出其更强的相似性度量能力与稳健性。
A trajectory similarity measure combining improved Word2Vec method and dynamic semantic map is proposed. The method learns the bayonet sequence correlation by the Word2Vec model with added destination constraint and establishes explicit connection with the destination; on the other hand, the dynamic semantic map can build the basis for the similarity measure of time and travel behavior dimension. The experimental results indicate that urban functional zones exhibit significant dynamic changes throughout the day. In the trajectory hierarchical clustering task, the average AC value of the method in this paper is 0.36 lower than that of the method chosen for comparison, further demonstrating its superior similarity measurement capability and robustness.
池娇 , 焦利民 , 董婷 , 等 , 2016 . 基于POI数据的城市功能区定量识别及其可视化 [J]. 测绘地理信息 , 41 ( 2 ): 68 - 73 .
江婧 , 张怀峰 , 皮德常 , 2019 . 基于卷积神经网络的移动对象目的地预测 [J]. 小型微型计算机系统 , 40 ( 12 ): 2519 - 2525 .
晋广印 , 赵旭俊 , 龚艺璇 , 2024 . 基于长短期记忆网络的移动轨迹目的地预测 [J]. 计算机工程与科学 , 46 ( 3 ): 525 - 534 .
李威 , 2018 . 基于历史轨迹的车辆类别预测 [D]. 山东 : 山东大学 .
罗月童 , 汪涛 , 杨梦男 , 等 , 2021 . 基于历史行车轨迹集的车辆行为可视分析方法 [J]. 计算机科学 , 48 ( 9 ): 86 - 94 .
吴晨昊 , 向隆刚 , 张叶廷 , 等 , 2023 . 基于地理空间感知型表征学习的轨迹相似度计算 [J]. 测绘学报 , 52 ( 4 ): 670 - 678 .
余丹青 , 邬群勇 , 姚江涛 , 等 , 2023 . 融合卷积、注意力和MLP的出租车目的地预测 [J]. 计算机工程与应用 , 59 ( 11 ): 302 - 311 .
余志 , 黄敏 , 何兆成 , 2022 . 道路交通信息物理系统(TCPS)概述 [EB/OL].( 2022-05-18 )[ 2025-06-03 ]. https://www.openits.cn/news/821.jhtml https://www.openits.cn/news/821.jhtml .
郑国强 , 乔宇昊 , 孙思民 , 2023 . 基于POI数据的城市功能区识别研究 [J]. 地理空间信息 , 21 ( 10 ): 58 - 61 .
周杭 , 樊红 , 2022 . 基于众源地理数据的城市功能区及其热点的识别研究 [J]. 武汉大学学报(工学版) , 55 ( 4 ): 417 - 426 .
CAO S , WU L , WU J , et al , 2022 . A spatio-temporal sequence-to-sequence network for traffic flow prediction [J]. Inf Sci , 610 : 185 - 203 .
DOROSTI A , ALESHEIKH A A , SHARIF M , 2024 . Measuring trajectory similarity based on the spatio-temporal properties of moving objects in road networks [J]. Information , 15 ( 1 ): 51 .
DU S , ZHANG H , XU H , et al , 2019 . To make the travel healthier:A new tourism personalized route recommendation algorithm [J]. J Ambient Intell Humaniz Comput , 10 ( 9 ): 3551 - 3562 .
FURTADO A S , KOPANAKI D , ALVARES L O , et al , 2016 . Multidimensional similarity measuring for semantic trajectories [J]. Trans GIS , 20 : 280 - 298 .
GAO S , JANOWICZ K , COUCLELIS H , 2017 . Extracting urban functional regions from points of interest and human activities on location-based social networks [J]. Trans GIS , 21 ( 3 ): 446 - 467 .
GUI Z , SUN Y , YANG L , et al , 2021 . LSI-LSTM: An attention-aware LSTM for real-time driving destination prediction by considering location semantics and location importance of trajectory points [J]. Neurocomputing , 440 : 72 - 88 .
KANG J , MA H , DUAN Z , et al , 2021 . Vehicle trajectory clustering in urban road network environment based on Doc2Vec model [C]// International Joint Conference on Neural Networks . Shenzhen, China : 1 - 8 .
LI X , ZHAO K , CONG G , et al , 2018 . Deep representation learning for trajectory similarity computation [C]// IEEE 34th International Conference on Data Engineering . Paris, France : 617 - 628 .
LI Y , LIU C , LI Y , 2022 . Identification of urban functional areas and their mixing degree using point of interest analyses [J]. Land , 11 ( 7 ): 996 .
LIANG M , LIU R W , LI S , et al , 2021 . An unsupervised learning method with convolutional auto-encoder for vessel trajectory similarity computation [J]. Ocean Eng , 225 : 108803 .
MIKOLOV T , CHEN K , CORRADO G , et al , 2013 . Efficient estimation of word representations in vector space [C]// International Conference on Learning Representations . Chicago, America : 1301 - 1313 .
SCHLOSSER F , BROCKMANN D , 2021 . Finding disease outbreak locations from human mobility data [J]. EPJ Data Sci , 10 ( 1 ): 1 - 17 .
SHANG S , CHEN L , WEI Z , et al , 2018 . Parallel trajectory similarity joins in spatial networks [J]. VLDB J , 27 ( 3 ): 395 - 420 .
XIA Y , WANG G Y , ZHANG X , et al , 2011 . Spatio-temporal similarity measure for network constrained trajectory data [J]. Int J Comput Intell Sys , 4 ( 5 ): 1070 - 1079 .
XUE A Y , ZHANG R , ZHENG Y , et al , 2013 . DesTeller: A system for destination prediction based on trajectories with privacy protection [J]. Proc VLDB Endow , 6 ( 12 ): 1198 - 1201 .
0
浏览量
9
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621
