2023
Yue, Yifan; Chen, Jun; Feng, Tao; Wang, Wei; Wang, Chunyang; Ma, Xinwei
In: Journal of Transportation Engineering, Part A: Systems, vol. 149, iss. 11, 2023.
Abstract | Links | BibTeX | Tags: Big data, classification algorithm, High-Speed Rail Station
@article{nokey,
title = {New Classification Scheme and Evolution Characteristics Analysis of High-Speed Railway Stations Using Large-Scale Mobile Phone Data: A Case Study in Jiangsu, China},
author = {Yue, Yifan and Chen, Jun and Feng, Tao and Wang, Wei and Wang, Chunyang and Ma, Xinwei},
url = {https://ascelibrary.org/doi/abs/10.1061/JTEPBS.TEENG-7855},
doi = {https://doi.org/10.1061/JTEPBS.TEENG-7855},
year = {2023},
date = {2023-09-06},
urldate = {2023-09-06},
journal = {Journal of Transportation Engineering, Part A: Systems},
volume = {149},
issue = {11},
abstract = {Effective management of the high-speed railways (HSR) system requires an in-depth understanding of the HSR stations in the network, e.g., the time-dependent volume distribution. The classification of HSR stations is the scientific basis for transport policymaking and land-use planning. Existing classification methods cannot meet the needs of temporal variation of passenger flow or the refined design and operation of HSR stations. This study adopts the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to classify HSR stations in different years. Using the data of Jiangsu Province, China, as an example, the time series of arrival and departure passenger flow at HSR stations are clustered via the DBSCAN algorithm, and the HSR stations are clustered into three classes. To determine the hierarchical structure of HSR stations representing the evolution of HSR networks, we use large-scale panel data obtained from mobile phone cellular data across years (July 1–14 from each of the years 2018, 2020, and 2021) to capture and analyze the spatial-temporal evolution characteristics of massive passenger flow at HSR stations. It is indicated that both HSR station hierarchy and passenger flow have the characteristics of spatial-temporal evolution across years, and the classification results are influenced by the geographical positions of cities and HSR layout. Accurate clustering of HSR stations via large-scale actual passenger flow data enables railway authorities and operators to identify critical nodes for efficient HSR network performance. The resulting classification would contribute to an in-depth understanding of the evolution characteristics of passenger flow in different years.},
keywords = {Big data, classification algorithm, High-Speed Rail Station},
pubstate = {published},
tppubtype = {article}
}
Effective management of the high-speed railways (HSR) system requires an in-depth understanding of the HSR stations in the network, e.g., the time-dependent volume distribution. The classification of HSR stations is the scientific basis for transport policymaking and land-use planning. Existing classification methods cannot meet the needs of temporal variation of passenger flow or the refined design and operation of HSR stations. This study adopts the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to classify HSR stations in different years. Using the data of Jiangsu Province, China, as an example, the time series of arrival and departure passenger flow at HSR stations are clustered via the DBSCAN algorithm, and the HSR stations are clustered into three classes. To determine the hierarchical structure of HSR stations representing the evolution of HSR networks, we use large-scale panel data obtained from mobile phone cellular data across years (July 1–14 from each of the years 2018, 2020, and 2021) to capture and analyze the spatial-temporal evolution characteristics of massive passenger flow at HSR stations. It is indicated that both HSR station hierarchy and passenger flow have the characteristics of spatial-temporal evolution across years, and the classification results are influenced by the geographical positions of cities and HSR layout. Accurate clustering of HSR stations via large-scale actual passenger flow data enables railway authorities and operators to identify critical nodes for efficient HSR network performance. The resulting classification would contribute to an in-depth understanding of the evolution characteristics of passenger flow in different years.
2016
Feng, T.; Timmermans, H. J. P.
Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data Journal Article
In: Transportation Planning and Technology, vol. 39, iss. 2, 2016, ISSN: 10290354.
Abstract | Links | BibTeX | Tags: activity-travel data, Bayesian network, classification algorithm, data imputation, decision tree, Global Positioning System (GPS), rules, travel survey
@article{Feng2016,
title = {Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data},
author = {T. Feng and H. J. P. Timmermans},
doi = {10.1080/03081060.2015.1127540},
issn = {10290354},
year = {2016},
date = {2016-01-01},
journal = {Transportation Planning and Technology},
volume = {39},
issue = {2},
abstract = {Global Positioning System (GPS) technologies have been increasingly considered as an alternative to traditional travel survey methods to collect activity-travel data. Algorithms applied to extract activity-travel patterns vary from informal ad-hoc decision rules to advanced machine learning methods and have different accuracy. This paper systematically compares the relative performance of different algorithms for the detection of transportation modes and activity episodes. In particular, naive Bayesian, Bayesian network, logistic regression, multilayer perceptron, support vector machine, decision table, and C4.5 algorithms are selected and compared for the same data according to their overall error rates and hit ratios. Results show that the Bayesian network has a better performance than the other algorithms in terms of the percentage correctly identified instances and Kappa values for both the training data and test data, in the sense that the Bayesian network is relatively efficient and generalizable in the context of GPS data imputation.},
keywords = {activity-travel data, Bayesian network, classification algorithm, data imputation, decision tree, Global Positioning System (GPS), rules, travel survey},
pubstate = {published},
tppubtype = {article}
}
Global Positioning System (GPS) technologies have been increasingly considered as an alternative to traditional travel survey methods to collect activity-travel data. Algorithms applied to extract activity-travel patterns vary from informal ad-hoc decision rules to advanced machine learning methods and have different accuracy. This paper systematically compares the relative performance of different algorithms for the detection of transportation modes and activity episodes. In particular, naive Bayesian, Bayesian network, logistic regression, multilayer perceptron, support vector machine, decision table, and C4.5 algorithms are selected and compared for the same data according to their overall error rates and hit ratios. Results show that the Bayesian network has a better performance than the other algorithms in terms of the percentage correctly identified instances and Kappa values for both the training data and test data, in the sense that the Bayesian network is relatively efficient and generalizable in the context of GPS data imputation.