Comparison of multivariate classification methods for contamination event detection in Water Distribution Systems

N. Oliker; Avi Ostfeld

doi:https://doi.org/10.1016/j.proeng.2014.02.140

Comparison of multivariate classification methods for contamination event detection in Water Distribution Systems

N. Oliker, Avi Ostfeld

Technion - Israel Institute of Technology

Research output: Contribution to journal › Conference article › peer-review

Abstract

This paper explores two applied classification models supplying decision support system for contamination event detection in Water Distribution Systems (WDS). The two models include an outlier's detection model and a following sequence analysis for the classification of event. The first model is an un-supervised minimum volume ellipsoid (MVE) and the second is a supervised support vector machine (SVM). The novelty of the two models is the multi-dimensional analysis of the data, differing from the parallel one-dimensional analysis that was conducted so far. The performance of the two models for the given problem is presented and compared.

Original language	English
Pages (from-to)	1271-1279
Number of pages	9
Journal	Procedia Engineering
Volume	70
DOIs	https://doi.org/10.1016/j.proeng.2014.02.140
State	Published - 2014
Event	12th International Conference on Computing and Control for the Water Industry, CCWI 2013 - Perugia, Italy Duration: 2 Sep 2013 → 4 Sep 2013

Keywords

Event detection
Minimum volume ellipsoid
Support vector machine
Water Distribution Systems
Water quality, water security

All Science Journal Classification (ASJC) codes

General Engineering

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

https://doi.org/10.1016/j.proeng.2014.02.140

Cite this

@article{3317e777b8524a5ca7bab7f20f2359b5,

title = "Comparison of multivariate classification methods for contamination event detection in Water Distribution Systems",

abstract = "This paper explores two applied classification models supplying decision support system for contamination event detection in Water Distribution Systems (WDS). The two models include an outlier's detection model and a following sequence analysis for the classification of event. The first model is an un-supervised minimum volume ellipsoid (MVE) and the second is a supervised support vector machine (SVM). The novelty of the two models is the multi-dimensional analysis of the data, differing from the parallel one-dimensional analysis that was conducted so far. The performance of the two models for the given problem is presented and compared.",

keywords = "Event detection, Minimum volume ellipsoid, Support vector machine, Water Distribution Systems, Water quality, water security",

author = "N. Oliker and Avi Ostfeld",

year = "2014",

doi = "https://doi.org/10.1016/j.proeng.2014.02.140",

language = "الإنجليزيّة",

volume = "70",

pages = "1271--1279",

journal = "Procedia Engineering",

issn = "1877-7058",

publisher = "Elsevier BV",

note = "12th International Conference on Computing and Control for the Water Industry, CCWI 2013 ; Conference date: 02-09-2013 Through 04-09-2013",

}

TY - JOUR

T1 - Comparison of multivariate classification methods for contamination event detection in Water Distribution Systems

AU - Oliker, N.

AU - Ostfeld, Avi

PY - 2014

Y1 - 2014

N2 - This paper explores two applied classification models supplying decision support system for contamination event detection in Water Distribution Systems (WDS). The two models include an outlier's detection model and a following sequence analysis for the classification of event. The first model is an un-supervised minimum volume ellipsoid (MVE) and the second is a supervised support vector machine (SVM). The novelty of the two models is the multi-dimensional analysis of the data, differing from the parallel one-dimensional analysis that was conducted so far. The performance of the two models for the given problem is presented and compared.

AB - This paper explores two applied classification models supplying decision support system for contamination event detection in Water Distribution Systems (WDS). The two models include an outlier's detection model and a following sequence analysis for the classification of event. The first model is an un-supervised minimum volume ellipsoid (MVE) and the second is a supervised support vector machine (SVM). The novelty of the two models is the multi-dimensional analysis of the data, differing from the parallel one-dimensional analysis that was conducted so far. The performance of the two models for the given problem is presented and compared.

KW - Event detection

KW - Minimum volume ellipsoid

KW - Support vector machine

KW - Water Distribution Systems

KW - Water quality, water security

UR - http://www.scopus.com/inward/record.url?scp=84899687718&partnerID=8YFLogxK

U2 - https://doi.org/10.1016/j.proeng.2014.02.140

DO - https://doi.org/10.1016/j.proeng.2014.02.140

M3 - مقالة من مؤنمر

SN - 1877-7058

VL - 70

SP - 1271

EP - 1279

JO - Procedia Engineering

JF - Procedia Engineering

T2 - 12th International Conference on Computing and Control for the Water Industry, CCWI 2013

Y2 - 2 September 2013 through 4 September 2013

ER -

Comparison of multivariate classification methods for contamination event detection in Water Distribution Systems

Abstract

Keywords

All Science Journal Classification (ASJC) codes

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this