Outlier Detection in Urban Traffic Flow Distributions

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

63 Downloads (Pure)

Abstract

Urban traffic data consists of observations like number and speed of cars or other vehicles at certain locations as measured by deployed sensors. These numbers can be interpreted as traffic flow which in turn relates to the capacity of streets and the demand of the traffic system. City planners are interested in studying the impact of various conditions on the traffic flow, leading to unusual patterns, i.e., outliers.
Existing approaches to outlier detection in urban traffic data take into account only individual flow values (i.e., an individual observation). This can be interesting for real time detection of sudden changes. Here, we face a different scenario: The city planners want to learn from historical data, how special circumstances (e.g., events or festivals) relate to unusual patterns in the traffic flow, in order to support improved planing of both, events and the layout of the traffic system.
Therefore, we propose to consider the sequence of traffic flow values observed within some time interval. Such flow sequences can be modeled as probability distributions of flows. We adapt an established outlier detection method, the local outlier factor (LOF), to handling flow distributions rather than individual observations. We apply the outlier detection online to extend the database with new flow distributions that are considered inliers. For the validation we consider a special case of our framework for comparison with state-of-the-art outlier detection on flows. In addition, a real case study on urban traffic flow data showcases that our method finds meaningful outliers in the traffic flow data.
Original languageEnglish
Title of host publication2018 IEEE International Conference on Data Mining
PublisherIEEE
Publication date31 Dec 2018
Pages935-940
ISBN (Print)978-1-5386-9160-1
ISBN (Electronic)978-1-5386-9159-5
DOIs
Publication statusPublished - 31 Dec 2018
EventIEEE International Conference on Data Mining - Singapore, Singapore
Duration: 17 Nov 201820 Nov 2018
http://icdm2018.org/

Conference

ConferenceIEEE International Conference on Data Mining
CountrySingapore
CitySingapore
Period17/11/201820/11/2018
Internet address

Fingerprint

Probability distributions
Railroad cars
Sensors

Keywords

  • Flow distributions
  • Outlier detection
  • Urban traffic data

Cite this

Djenouri, Y., Zimek, A., & Chiarandini, M. (2018). Outlier Detection in Urban Traffic Flow Distributions. In 2018 IEEE International Conference on Data Mining (pp. 935-940). IEEE. https://doi.org/10.1109/ICDM.2018.00114
Djenouri, Youcef ; Zimek, Arthur ; Chiarandini, Marco. / Outlier Detection in Urban Traffic Flow Distributions. 2018 IEEE International Conference on Data Mining. IEEE, 2018. pp. 935-940
@inproceedings{9c5c0f67da6d489c9a696f56516cece0,
title = "Outlier Detection in Urban Traffic Flow Distributions",
abstract = "Urban traffic data consists of observations like number and speed of cars or other vehicles at certain locations as measured by deployed sensors. These numbers can be interpreted as traffic flow which in turn relates to the capacity of streets and the demand of the traffic system. City planners are interested in studying the impact of various conditions on the traffic flow, leading to unusual patterns, i.e., outliers.Existing approaches to outlier detection in urban traffic data take into account only individual flow values (i.e., an individual observation). This can be interesting for real time detection of sudden changes. Here, we face a different scenario: The city planners want to learn from historical data, how special circumstances (e.g., events or festivals) relate to unusual patterns in the traffic flow, in order to support improved planing of both, events and the layout of the traffic system.Therefore, we propose to consider the sequence of traffic flow values observed within some time interval. Such flow sequences can be modeled as probability distributions of flows. We adapt an established outlier detection method, the local outlier factor (LOF), to handling flow distributions rather than individual observations. We apply the outlier detection online to extend the database with new flow distributions that are considered inliers. For the validation we consider a special case of our framework for comparison with state-of-the-art outlier detection on flows. In addition, a real case study on urban traffic flow data showcases that our method finds meaningful outliers in the traffic flow data.",
keywords = "Flow distributions, Outlier detection, Urban traffic data",
author = "Youcef Djenouri and Arthur Zimek and Marco Chiarandini",
note = "Accepted Manuscript {\circledC} 2018 IEEE",
year = "2018",
month = "12",
day = "31",
doi = "10.1109/ICDM.2018.00114",
language = "English",
isbn = "978-1-5386-9160-1",
pages = "935--940",
booktitle = "2018 IEEE International Conference on Data Mining",
publisher = "IEEE",
address = "United States",

}

Djenouri, Y, Zimek, A & Chiarandini, M 2018, Outlier Detection in Urban Traffic Flow Distributions. in 2018 IEEE International Conference on Data Mining. IEEE, pp. 935-940, IEEE International Conference on Data Mining, Singapore, Singapore, 17/11/2018. https://doi.org/10.1109/ICDM.2018.00114

Outlier Detection in Urban Traffic Flow Distributions. / Djenouri, Youcef; Zimek, Arthur; Chiarandini, Marco.

2018 IEEE International Conference on Data Mining. IEEE, 2018. p. 935-940.

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - Outlier Detection in Urban Traffic Flow Distributions

AU - Djenouri, Youcef

AU - Zimek, Arthur

AU - Chiarandini, Marco

N1 - Accepted Manuscript © 2018 IEEE

PY - 2018/12/31

Y1 - 2018/12/31

N2 - Urban traffic data consists of observations like number and speed of cars or other vehicles at certain locations as measured by deployed sensors. These numbers can be interpreted as traffic flow which in turn relates to the capacity of streets and the demand of the traffic system. City planners are interested in studying the impact of various conditions on the traffic flow, leading to unusual patterns, i.e., outliers.Existing approaches to outlier detection in urban traffic data take into account only individual flow values (i.e., an individual observation). This can be interesting for real time detection of sudden changes. Here, we face a different scenario: The city planners want to learn from historical data, how special circumstances (e.g., events or festivals) relate to unusual patterns in the traffic flow, in order to support improved planing of both, events and the layout of the traffic system.Therefore, we propose to consider the sequence of traffic flow values observed within some time interval. Such flow sequences can be modeled as probability distributions of flows. We adapt an established outlier detection method, the local outlier factor (LOF), to handling flow distributions rather than individual observations. We apply the outlier detection online to extend the database with new flow distributions that are considered inliers. For the validation we consider a special case of our framework for comparison with state-of-the-art outlier detection on flows. In addition, a real case study on urban traffic flow data showcases that our method finds meaningful outliers in the traffic flow data.

AB - Urban traffic data consists of observations like number and speed of cars or other vehicles at certain locations as measured by deployed sensors. These numbers can be interpreted as traffic flow which in turn relates to the capacity of streets and the demand of the traffic system. City planners are interested in studying the impact of various conditions on the traffic flow, leading to unusual patterns, i.e., outliers.Existing approaches to outlier detection in urban traffic data take into account only individual flow values (i.e., an individual observation). This can be interesting for real time detection of sudden changes. Here, we face a different scenario: The city planners want to learn from historical data, how special circumstances (e.g., events or festivals) relate to unusual patterns in the traffic flow, in order to support improved planing of both, events and the layout of the traffic system.Therefore, we propose to consider the sequence of traffic flow values observed within some time interval. Such flow sequences can be modeled as probability distributions of flows. We adapt an established outlier detection method, the local outlier factor (LOF), to handling flow distributions rather than individual observations. We apply the outlier detection online to extend the database with new flow distributions that are considered inliers. For the validation we consider a special case of our framework for comparison with state-of-the-art outlier detection on flows. In addition, a real case study on urban traffic flow data showcases that our method finds meaningful outliers in the traffic flow data.

KW - Flow distributions

KW - Outlier detection

KW - Urban traffic data

U2 - 10.1109/ICDM.2018.00114

DO - 10.1109/ICDM.2018.00114

M3 - Article in proceedings

SN - 978-1-5386-9160-1

SP - 935

EP - 940

BT - 2018 IEEE International Conference on Data Mining

PB - IEEE

ER -

Djenouri Y, Zimek A, Chiarandini M. Outlier Detection in Urban Traffic Flow Distributions. In 2018 IEEE International Conference on Data Mining. IEEE. 2018. p. 935-940 https://doi.org/10.1109/ICDM.2018.00114