Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data

Research output: Contribution to journalArticleAcademicpeer-review

16 Citations (Scopus)
183 Downloads (Pure)

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.

Original languageEnglish
Pages (from-to)180-194
Number of pages15
JournalTransportation Planning and Technology
Volume39
Issue number2
DOIs
Publication statusPublished - 17 Feb 2016

Fingerprint

Global positioning system
Bayesian networks
GPS
travel
Decision tables
Multilayer neural networks
survey method
Support vector machines
Learning systems
Logistics
learning method
performance
logistics
detection
comparison
mode of transportation
regression
decision

Keywords

  • activity-travel data
  • Bayesian network
  • classification algorithm
  • data imputation
  • decision tree
  • Global Positioning System (GPS)
  • rules
  • Travel survey

Cite this

@article{3309692dd0bb4f32b546a83036455c54,
title = "Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data",
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",
author = "T. Feng and H.J.P. Timmermans",
year = "2016",
month = "2",
day = "17",
doi = "10.1080/03081060.2015.1127540",
language = "English",
volume = "39",
pages = "180--194",
journal = "Transportation Planning and Technology",
issn = "0308-1060",
publisher = "Taylor and Francis Ltd.",
number = "2",

}

Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data. / Feng, T.; Timmermans, H.J.P.

In: Transportation Planning and Technology, Vol. 39, No. 2, 17.02.2016, p. 180-194.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data

AU - Feng, T.

AU - Timmermans, H.J.P.

PY - 2016/2/17

Y1 - 2016/2/17

N2 - 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.

AB - 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.

KW - activity-travel data

KW - Bayesian network

KW - classification algorithm

KW - data imputation

KW - decision tree

KW - Global Positioning System (GPS)

KW - rules

KW - Travel survey

UR - http://www.scopus.com/inward/record.url?scp=84958848203&partnerID=8YFLogxK

U2 - 10.1080/03081060.2015.1127540

DO - 10.1080/03081060.2015.1127540

M3 - Article

AN - SCOPUS:84958848203

VL - 39

SP - 180

EP - 194

JO - Transportation Planning and Technology

JF - Transportation Planning and Technology

SN - 0308-1060

IS - 2

ER -