Outlier detection and comparison of origin-destination flows using data depth

Myeong Hun Jeong, Junjun Yin, Shaowen Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Advances in location-aware technology have resulted in massive trajectory data. Origin-destination (OD) trajectories provide rich information on urban flow and transport demand. This study describes a new method for detecting OD flows outliers and conducting hypothesis testing between two OD flow datasets in terms of the variations of spatial extent, that is, spread. The proposed method is based on data depth, which measures the centrality and outlyingness of a point with respect to a given dataset in Rd. Based on the center-outward ordering property, the proposed method analyzes the underlying characteristics of OD flows, such as location, outlyingness, and spread. The ability of the method to detect OD anomalies is compared with that of the Mahalanobis distance approach, and an F-test is used to verify the difference in scale. Empirical evaluation has demonstrated that our method effectively identifies OD flows outliers in an interactive way. Furthermore, the method can provide new perspectives such as spatial extent by considering the overall structure of data when comparing two different OD flows in terms of scale.

Original languageEnglish (US)
Title of host publication10th International Conference on Geographic Information Science, GIScience 2018
EditorsAmy L. Griffin, Stephan Winter, Monika Sester
PublisherSchloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing
ISBN (Print)9783959770835
DOIs
StatePublished - Aug 1 2018
Event10th International Conference on Geographic Information Science, GIScience 2018 - Melbourne, Australia
Duration: Aug 28 2018Aug 31 2018

Publication series

NameLeibniz International Proceedings in Informatics, LIPIcs
Volume114
ISSN (Print)1868-8969

Other

Other10th International Conference on Geographic Information Science, GIScience 2018
CountryAustralia
CityMelbourne
Period8/28/188/31/18

Fingerprint

Trajectories
Testing

All Science Journal Classification (ASJC) codes

  • Software

Cite this

Jeong, M. H., Yin, J., & Wang, S. (2018). Outlier detection and comparison of origin-destination flows using data depth. In A. L. Griffin, S. Winter, & M. Sester (Eds.), 10th International Conference on Geographic Information Science, GIScience 2018 (Leibniz International Proceedings in Informatics, LIPIcs; Vol. 114). Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing. https://doi.org/10.4230/LIPIcs.GIScience.2018.6
Jeong, Myeong Hun ; Yin, Junjun ; Wang, Shaowen. / Outlier detection and comparison of origin-destination flows using data depth. 10th International Conference on Geographic Information Science, GIScience 2018. editor / Amy L. Griffin ; Stephan Winter ; Monika Sester. Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing, 2018. (Leibniz International Proceedings in Informatics, LIPIcs).
@inproceedings{49eac963b898495db85f4502c94239b6,
title = "Outlier detection and comparison of origin-destination flows using data depth",
abstract = "Advances in location-aware technology have resulted in massive trajectory data. Origin-destination (OD) trajectories provide rich information on urban flow and transport demand. This study describes a new method for detecting OD flows outliers and conducting hypothesis testing between two OD flow datasets in terms of the variations of spatial extent, that is, spread. The proposed method is based on data depth, which measures the centrality and outlyingness of a point with respect to a given dataset in Rd. Based on the center-outward ordering property, the proposed method analyzes the underlying characteristics of OD flows, such as location, outlyingness, and spread. The ability of the method to detect OD anomalies is compared with that of the Mahalanobis distance approach, and an F-test is used to verify the difference in scale. Empirical evaluation has demonstrated that our method effectively identifies OD flows outliers in an interactive way. Furthermore, the method can provide new perspectives such as spatial extent by considering the overall structure of data when comparing two different OD flows in terms of scale.",
author = "Jeong, {Myeong Hun} and Junjun Yin and Shaowen Wang",
year = "2018",
month = "8",
day = "1",
doi = "10.4230/LIPIcs.GIScience.2018.6",
language = "English (US)",
isbn = "9783959770835",
series = "Leibniz International Proceedings in Informatics, LIPIcs",
publisher = "Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing",
editor = "Griffin, {Amy L.} and Stephan Winter and Monika Sester",
booktitle = "10th International Conference on Geographic Information Science, GIScience 2018",

}

Jeong, MH, Yin, J & Wang, S 2018, Outlier detection and comparison of origin-destination flows using data depth. in AL Griffin, S Winter & M Sester (eds), 10th International Conference on Geographic Information Science, GIScience 2018. Leibniz International Proceedings in Informatics, LIPIcs, vol. 114, Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing, 10th International Conference on Geographic Information Science, GIScience 2018, Melbourne, Australia, 8/28/18. https://doi.org/10.4230/LIPIcs.GIScience.2018.6

Outlier detection and comparison of origin-destination flows using data depth. / Jeong, Myeong Hun; Yin, Junjun; Wang, Shaowen.

10th International Conference on Geographic Information Science, GIScience 2018. ed. / Amy L. Griffin; Stephan Winter; Monika Sester. Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing, 2018. (Leibniz International Proceedings in Informatics, LIPIcs; Vol. 114).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Outlier detection and comparison of origin-destination flows using data depth

AU - Jeong, Myeong Hun

AU - Yin, Junjun

AU - Wang, Shaowen

PY - 2018/8/1

Y1 - 2018/8/1

N2 - Advances in location-aware technology have resulted in massive trajectory data. Origin-destination (OD) trajectories provide rich information on urban flow and transport demand. This study describes a new method for detecting OD flows outliers and conducting hypothesis testing between two OD flow datasets in terms of the variations of spatial extent, that is, spread. The proposed method is based on data depth, which measures the centrality and outlyingness of a point with respect to a given dataset in Rd. Based on the center-outward ordering property, the proposed method analyzes the underlying characteristics of OD flows, such as location, outlyingness, and spread. The ability of the method to detect OD anomalies is compared with that of the Mahalanobis distance approach, and an F-test is used to verify the difference in scale. Empirical evaluation has demonstrated that our method effectively identifies OD flows outliers in an interactive way. Furthermore, the method can provide new perspectives such as spatial extent by considering the overall structure of data when comparing two different OD flows in terms of scale.

AB - Advances in location-aware technology have resulted in massive trajectory data. Origin-destination (OD) trajectories provide rich information on urban flow and transport demand. This study describes a new method for detecting OD flows outliers and conducting hypothesis testing between two OD flow datasets in terms of the variations of spatial extent, that is, spread. The proposed method is based on data depth, which measures the centrality and outlyingness of a point with respect to a given dataset in Rd. Based on the center-outward ordering property, the proposed method analyzes the underlying characteristics of OD flows, such as location, outlyingness, and spread. The ability of the method to detect OD anomalies is compared with that of the Mahalanobis distance approach, and an F-test is used to verify the difference in scale. Empirical evaluation has demonstrated that our method effectively identifies OD flows outliers in an interactive way. Furthermore, the method can provide new perspectives such as spatial extent by considering the overall structure of data when comparing two different OD flows in terms of scale.

UR - http://www.scopus.com/inward/record.url?scp=85051360626&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051360626&partnerID=8YFLogxK

U2 - 10.4230/LIPIcs.GIScience.2018.6

DO - 10.4230/LIPIcs.GIScience.2018.6

M3 - Conference contribution

SN - 9783959770835

T3 - Leibniz International Proceedings in Informatics, LIPIcs

BT - 10th International Conference on Geographic Information Science, GIScience 2018

A2 - Griffin, Amy L.

A2 - Winter, Stephan

A2 - Sester, Monika

PB - Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing

ER -

Jeong MH, Yin J, Wang S. Outlier detection and comparison of origin-destination flows using data depth. In Griffin AL, Winter S, Sester M, editors, 10th International Conference on Geographic Information Science, GIScience 2018. Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing. 2018. (Leibniz International Proceedings in Informatics, LIPIcs). https://doi.org/10.4230/LIPIcs.GIScience.2018.6