Contextual spatial outlier detection with metric learning

Guanjie Zheng, Susan Louise Brantley, Thomas Claude Yves Lauvaux, Zhenhui Li

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

8 Citations (Scopus)

Abstract

Hydraulic fracturing (or "fracking") is a revolutionary well stimulation technique for shale gas extraction, but has spawned controversy in environmental contamination. If methane from gas wells leaks extensively this greenhouse gas can impact drinking water wells and enhance global warming. Our work is motivated by this heated debate on environmental issue and focuses on general data analytical techniques to detect anomalous spatial data samples (e.g., water samples related to potential leakages). Specifically, we propose a spatial outlier detection method based on contextual neighbors. Different from existing work, our approach utilizes both spatial attributes and non-spatial contextual attributes to define neighbors. We further use robust metric learning to combine different contextual attributes in order to find meaningful neighbors. Our technique can be applied to any spatial dataset. Extensive experimental results on five real-world datasets demonstrate the effectiveness of our approach. We also show some interesting case studies, including one case linking to leakage of a gas well.

Original languageEnglish (US)
Title of host publicationKDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Pages2161-2170
Number of pages10
ISBN (Electronic)9781450348874
DOIs
StatePublished - Aug 13 2017
Event23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2017 - Halifax, Canada
Duration: Aug 13 2017Aug 17 2017

Publication series

NameProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
VolumePart F129685

Other

Other23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2017
CountryCanada
CityHalifax
Period8/13/178/17/17

Fingerprint

Hydraulic fracturing
Well stimulation
Water wells
Leakage (fluid)
Global warming
Gases
Greenhouse gases
Potable water
Methane
Contamination
Water
Shale gas

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems

Cite this

Zheng, G., Brantley, S. L., Lauvaux, T. C. Y., & Li, Z. (2017). Contextual spatial outlier detection with metric learning. In KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 2161-2170). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; Vol. Part F129685). Association for Computing Machinery. https://doi.org/10.1145/3097983.3098143
Zheng, Guanjie ; Brantley, Susan Louise ; Lauvaux, Thomas Claude Yves ; Li, Zhenhui. / Contextual spatial outlier detection with metric learning. KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2017. pp. 2161-2170 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).
@inproceedings{450c58f3517c472b89f17a5a1978b60c,
title = "Contextual spatial outlier detection with metric learning",
abstract = "Hydraulic fracturing (or {"}fracking{"}) is a revolutionary well stimulation technique for shale gas extraction, but has spawned controversy in environmental contamination. If methane from gas wells leaks extensively this greenhouse gas can impact drinking water wells and enhance global warming. Our work is motivated by this heated debate on environmental issue and focuses on general data analytical techniques to detect anomalous spatial data samples (e.g., water samples related to potential leakages). Specifically, we propose a spatial outlier detection method based on contextual neighbors. Different from existing work, our approach utilizes both spatial attributes and non-spatial contextual attributes to define neighbors. We further use robust metric learning to combine different contextual attributes in order to find meaningful neighbors. Our technique can be applied to any spatial dataset. Extensive experimental results on five real-world datasets demonstrate the effectiveness of our approach. We also show some interesting case studies, including one case linking to leakage of a gas well.",
author = "Guanjie Zheng and Brantley, {Susan Louise} and Lauvaux, {Thomas Claude Yves} and Zhenhui Li",
year = "2017",
month = "8",
day = "13",
doi = "10.1145/3097983.3098143",
language = "English (US)",
series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",
publisher = "Association for Computing Machinery",
pages = "2161--2170",
booktitle = "KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

Zheng, G, Brantley, SL, Lauvaux, TCY & Li, Z 2017, Contextual spatial outlier detection with metric learning. in KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, vol. Part F129685, Association for Computing Machinery, pp. 2161-2170, 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2017, Halifax, Canada, 8/13/17. https://doi.org/10.1145/3097983.3098143

Contextual spatial outlier detection with metric learning. / Zheng, Guanjie; Brantley, Susan Louise; Lauvaux, Thomas Claude Yves; Li, Zhenhui.

KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2017. p. 2161-2170 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; Vol. Part F129685).

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

TY - GEN

T1 - Contextual spatial outlier detection with metric learning

AU - Zheng, Guanjie

AU - Brantley, Susan Louise

AU - Lauvaux, Thomas Claude Yves

AU - Li, Zhenhui

PY - 2017/8/13

Y1 - 2017/8/13

N2 - Hydraulic fracturing (or "fracking") is a revolutionary well stimulation technique for shale gas extraction, but has spawned controversy in environmental contamination. If methane from gas wells leaks extensively this greenhouse gas can impact drinking water wells and enhance global warming. Our work is motivated by this heated debate on environmental issue and focuses on general data analytical techniques to detect anomalous spatial data samples (e.g., water samples related to potential leakages). Specifically, we propose a spatial outlier detection method based on contextual neighbors. Different from existing work, our approach utilizes both spatial attributes and non-spatial contextual attributes to define neighbors. We further use robust metric learning to combine different contextual attributes in order to find meaningful neighbors. Our technique can be applied to any spatial dataset. Extensive experimental results on five real-world datasets demonstrate the effectiveness of our approach. We also show some interesting case studies, including one case linking to leakage of a gas well.

AB - Hydraulic fracturing (or "fracking") is a revolutionary well stimulation technique for shale gas extraction, but has spawned controversy in environmental contamination. If methane from gas wells leaks extensively this greenhouse gas can impact drinking water wells and enhance global warming. Our work is motivated by this heated debate on environmental issue and focuses on general data analytical techniques to detect anomalous spatial data samples (e.g., water samples related to potential leakages). Specifically, we propose a spatial outlier detection method based on contextual neighbors. Different from existing work, our approach utilizes both spatial attributes and non-spatial contextual attributes to define neighbors. We further use robust metric learning to combine different contextual attributes in order to find meaningful neighbors. Our technique can be applied to any spatial dataset. Extensive experimental results on five real-world datasets demonstrate the effectiveness of our approach. We also show some interesting case studies, including one case linking to leakage of a gas well.

UR - http://www.scopus.com/inward/record.url?scp=85029047552&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029047552&partnerID=8YFLogxK

U2 - 10.1145/3097983.3098143

DO - 10.1145/3097983.3098143

M3 - Conference contribution

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 2161

EP - 2170

BT - KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -

Zheng G, Brantley SL, Lauvaux TCY, Li Z. Contextual spatial outlier detection with metric learning. In KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery. 2017. p. 2161-2170. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). https://doi.org/10.1145/3097983.3098143