Inferring the root cause in road traffic anomalies

Sanjay Chawla, Yu Zheng, Jiafeng Hu

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

106 Citations (Scopus)

Abstract

We propose a novel two-step mining and optimization framework for inferring the root cause of anomalies that appear in road traffic data. We model road traffic as a time-dependent flow on a network formed by partitioning a city into regions bounded by major roads. In the first step we identify link anomalies based on their deviation from their historical traffic profile. However, link anomalies on their own shed very little light on what caused them to be anomalous. In the second step we take a generative approach by modeling the flow in a network in terms of the origin-destination (OD) matrix which physically relates the latent flow between origin and destination and the observable flow on the links. The key insight is that instead of using all of link traffic as the observable vector we only use the link anomaly vector. By solving an L 1 inverse problem we infer the routes (the origin-destination pairs) which gave rise to the link anomalies. Experiments on a very large GPS data set consisting on nearly eight hundred million data points demonstrate that we can discover routes which can clearly explain the appearance of link anomalies. The use of optimization techniques to explain observable anomalies in a generative fashion is, to the best of our knowledge, entirely novel.

Original languageEnglish
Title of host publicationProceedings - IEEE International Conference on Data Mining, ICDM
Pages141-150
Number of pages10
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event12th IEEE International Conference on Data Mining, ICDM 2012 - Brussels
Duration: 10 Dec 201213 Dec 2012

Other

Other12th IEEE International Conference on Data Mining, ICDM 2012
CityBrussels
Period10/12/1213/12/12

Fingerprint

Inverse problems
Telecommunication traffic
Telecommunication links
Global positioning system
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Chawla, S., Zheng, Y., & Hu, J. (2012). Inferring the root cause in road traffic anomalies. In Proceedings - IEEE International Conference on Data Mining, ICDM (pp. 141-150). [6413908] https://doi.org/10.1109/ICDM.2012.104

Inferring the root cause in road traffic anomalies. / Chawla, Sanjay; Zheng, Yu; Hu, Jiafeng.

Proceedings - IEEE International Conference on Data Mining, ICDM. 2012. p. 141-150 6413908.

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

Chawla, S, Zheng, Y & Hu, J 2012, Inferring the root cause in road traffic anomalies. in Proceedings - IEEE International Conference on Data Mining, ICDM., 6413908, pp. 141-150, 12th IEEE International Conference on Data Mining, ICDM 2012, Brussels, 10/12/12. https://doi.org/10.1109/ICDM.2012.104
Chawla S, Zheng Y, Hu J. Inferring the root cause in road traffic anomalies. In Proceedings - IEEE International Conference on Data Mining, ICDM. 2012. p. 141-150. 6413908 https://doi.org/10.1109/ICDM.2012.104
Chawla, Sanjay ; Zheng, Yu ; Hu, Jiafeng. / Inferring the root cause in road traffic anomalies. Proceedings - IEEE International Conference on Data Mining, ICDM. 2012. pp. 141-150
@inproceedings{7d3e54a30d634451b9b7e07ed491d9ea,
title = "Inferring the root cause in road traffic anomalies",
abstract = "We propose a novel two-step mining and optimization framework for inferring the root cause of anomalies that appear in road traffic data. We model road traffic as a time-dependent flow on a network formed by partitioning a city into regions bounded by major roads. In the first step we identify link anomalies based on their deviation from their historical traffic profile. However, link anomalies on their own shed very little light on what caused them to be anomalous. In the second step we take a generative approach by modeling the flow in a network in terms of the origin-destination (OD) matrix which physically relates the latent flow between origin and destination and the observable flow on the links. The key insight is that instead of using all of link traffic as the observable vector we only use the link anomaly vector. By solving an L 1 inverse problem we infer the routes (the origin-destination pairs) which gave rise to the link anomalies. Experiments on a very large GPS data set consisting on nearly eight hundred million data points demonstrate that we can discover routes which can clearly explain the appearance of link anomalies. The use of optimization techniques to explain observable anomalies in a generative fashion is, to the best of our knowledge, entirely novel.",
author = "Sanjay Chawla and Yu Zheng and Jiafeng Hu",
year = "2012",
doi = "10.1109/ICDM.2012.104",
language = "English",
isbn = "9780769549057",
pages = "141--150",
booktitle = "Proceedings - IEEE International Conference on Data Mining, ICDM",

}

TY - GEN

T1 - Inferring the root cause in road traffic anomalies

AU - Chawla, Sanjay

AU - Zheng, Yu

AU - Hu, Jiafeng

PY - 2012

Y1 - 2012

N2 - We propose a novel two-step mining and optimization framework for inferring the root cause of anomalies that appear in road traffic data. We model road traffic as a time-dependent flow on a network formed by partitioning a city into regions bounded by major roads. In the first step we identify link anomalies based on their deviation from their historical traffic profile. However, link anomalies on their own shed very little light on what caused them to be anomalous. In the second step we take a generative approach by modeling the flow in a network in terms of the origin-destination (OD) matrix which physically relates the latent flow between origin and destination and the observable flow on the links. The key insight is that instead of using all of link traffic as the observable vector we only use the link anomaly vector. By solving an L 1 inverse problem we infer the routes (the origin-destination pairs) which gave rise to the link anomalies. Experiments on a very large GPS data set consisting on nearly eight hundred million data points demonstrate that we can discover routes which can clearly explain the appearance of link anomalies. The use of optimization techniques to explain observable anomalies in a generative fashion is, to the best of our knowledge, entirely novel.

AB - We propose a novel two-step mining and optimization framework for inferring the root cause of anomalies that appear in road traffic data. We model road traffic as a time-dependent flow on a network formed by partitioning a city into regions bounded by major roads. In the first step we identify link anomalies based on their deviation from their historical traffic profile. However, link anomalies on their own shed very little light on what caused them to be anomalous. In the second step we take a generative approach by modeling the flow in a network in terms of the origin-destination (OD) matrix which physically relates the latent flow between origin and destination and the observable flow on the links. The key insight is that instead of using all of link traffic as the observable vector we only use the link anomaly vector. By solving an L 1 inverse problem we infer the routes (the origin-destination pairs) which gave rise to the link anomalies. Experiments on a very large GPS data set consisting on nearly eight hundred million data points demonstrate that we can discover routes which can clearly explain the appearance of link anomalies. The use of optimization techniques to explain observable anomalies in a generative fashion is, to the best of our knowledge, entirely novel.

UR - http://www.scopus.com/inward/record.url?scp=84874085057&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874085057&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2012.104

DO - 10.1109/ICDM.2012.104

M3 - Conference contribution

SN - 9780769549057

SP - 141

EP - 150

BT - Proceedings - IEEE International Conference on Data Mining, ICDM

ER -