Trans2Vec:耦合车辆移动模式的大规模城市路网表示学习框架

doi:10.13474/j.cnki.11-2246.2023.0075

测绘通报 ›› 2023, Vol. 0 ›› Issue (3): 74-78.doi: 10.13474/j.cnki.11-2246.2023.0075

Trans2Vec:耦合车辆移动模式的大规模城市路网表示学习框架

邓敏, 储国威, 谌恺祺, 石岩

中南大学, 湖南长沙 410083

收稿日期:2022-03-15 发布日期:2023-04-04
通讯作者: 储国威。E-mail:195011043@csu.edu.cn
作者简介:邓敏(1974-),男,博士,教授,研究方向为时空大数据挖掘。E-mail:dengmin@csu.edu.cn
基金资助:
国家自然科学基金（42171459；41730105）

Trans2Vec: a representation learning framework for large-scale urban road networks coupled with vehicle movement patterns

DENG Min, CHU Guowei, CHEN Kaiqi, SHI Yan

Center South University, Changsha 410083, China

Received:2022-03-15 Published:2023-04-04

摘要/Abstract

摘要： 高效率实现城市路网信息的完备向量化表征，是嵌入各类深度学习模型以智能化解决下游交通任务的重要数据工程。现有表示学习方法难以有效耦合城市路网固有的拓扑结构信息与承载的转移模式信息，无法精确匹配下游交通任务的特征需求，且在面对大规模城市路网时，存在算法资源占用较高的现实问题。针对该问题，本文发展了一种多视图游走的路段表征向量学习方法，以车辆移动模式为核心，融合拓扑结构进行异质游走，实现大规模路网完备、高效的信息表达；以深圳市为试验区域，通过全量城市路网与实际车辆行程数据验证了该方法的有效性与先进性。

关键词: Trans2Vec, 城市路网, 车辆移动模式, 表示学习

Abstract: Efficient learning of representation of urban road network information is an important project for embedding various deep learning models to intelligently solve downstream traffic tasks. However, the existing learning framework are difficult to effectively couple the topology information of the urban road network and the transfer mode information carried, and cannot accurately match the characteristic requirements of downstream traffic tasks. In this reason, this paper develops a multi-view roaming segment representation vector learning framework, which takes the vehicle movement pattern as the core, integrates the topology structure for heterogeneous walk, and realizes the complete and efficient information embedding of large-scale road network. In the experimental part, taking Shenzhen as the case area, the effectiveness and advanced nature of the method in this paper are verified through the full urban road network and actual vehicle travel data.

Key words: Trans2Vec, urban road network, vehicle movement patterns, representation learning

中图分类号:

P208

邓敏, 储国威, 谌恺祺, 石岩. Trans2Vec:耦合车辆移动模式的大规模城市路网表示学习框架[J]. 测绘通报, 2023, 0(3): 74-78.

DENG Min, CHU Guowei, CHEN Kaiqi, SHI Yan. Trans2Vec: a representation learning framework for large-scale urban road networks coupled with vehicle movement patterns[J]. Bulletin of Surveying and Mapping, 2023, 0(3): 74-78.

参考文献

[1] DIMITRAKOPOULOS G, DEMESTICHAS P. Intelligent transportation systems[J]. IEEE Vehicular Technology Magazine, 2010, 5(1):77-84.
[2] WANG Mengxiang, LEE W C, FU Taoyang, et al. On representation learning for road networks[J]. ACM Transactions on Intelligent Systems and Technology, 2021, 12(1):1-27.
[3] JEPSEN T S, JENSEN C S, NIELSEN T D, et al. On network embedding for machine learning on road networks:a case study on the Danish road network[C]//Proceedings of 2018 IEEE International Conference on Big Data.Seattle, WA, USA:IEEE, 2018.
[4] LIU Kang, GAO Song, QIU Peiyuan, et al. Road2Vec:measuring traffic interactions in urban road system from massive travel routes[J]. ISPRS International Journal of Geo-Information, 2017, 6(11):321.
[5] WANG Hongjian, TANG Xianfeng, KUO Y H, et al. A simple baseline for travel time estimation using large-scale trip data[J]. ACM Transactions on Intelligent Systems and Technology, 2019, 10(2):1-22.
[6] WANG Mengxiang, LEE W C, FU Taoyang, et al. Learning embeddings of intersections on road networks[C]//Proceedings of the 27th ACM SigSpatial International Conference on Advances in Geographic Information Systems. New York:ACM, 2019:309-318.
[7] CUI Peng, WANG Xiao, PEI Jian, et al. A survey on network embedding[J]. IEEE Transactions on Knowledge and Data Engineering, 2019, 31(5):833-852.
[8] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323(6088):533-536.
[9] PEROZZI B, AL-RFOU R, SKIENA S. DeepWalk:online learning of social representations[C]//Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining. New York:ACM, 2014:701-710.
[10] GROVER A, LESKOVEC J.Node2vec:scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:ACM, 2016:855-864.
[11] CLAUSET A, NEWMAN M J, MOORE C. Finding community structure in very large networks[J]. Physical Review E, Statistical, Nonlinear, and Soft Matter Physics, 2004, 70(6):066111.
[12] CHEN Kaiqi, DENG Min, SHI Yan. A temporal directed graph convolution network for traffic forecasting using taxi trajectory data[J]. ISPRS International Journal of Geo-Information, 2021, 10(9):624.
[13] LI A Q, AHMED A, RAVI S, et al.Reducing the sampling complexity of topic models[C]//Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining. New York:ACM, 2014:891-900.
[14] WANG Zheng, FU Kun, YE Jieping. Learning to estimate the travel time[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery&Data Mining.New York:ACM Press, 2018:858-866.

Trans2Vec:耦合车辆移动模式的大规模城市路网表示学习框架

Trans2Vec: a representation learning framework for large-scale urban road networks coupled with vehicle movement patterns

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价