Abstract

Detection of changes to multivariate patterns is an important topic in a number of different domains. Modern data sets often include categorical and numerical data and potentially complex in-control regions. Given a flexible, robust decision rule for this environment that signals based on an individual observation vector, an important issue is how to extend the rule to incorporate time-based information. A decision rule can be learned to detect shifts through artificial data that transforms the problem to one of supervised learning. Then class probability ratios are derived from a relationship to likelihood ratios to form the basis for time-weighted updates of the monitoring scheme.

Original languageEnglish (US)
Pages (from-to)67-81
Number of pages15
JournalAnnals of Operations Research
Volume174
Issue number1
DOIs
StatePublished - Feb 2010

Fingerprint

Decision rules
Likelihood ratio
Monitoring

Keywords

  • Multivariate patterns
  • Supervised learning
  • Time-based detection of changes

ASJC Scopus subject areas

  • Management Science and Operations Research
  • Decision Sciences(all)

Cite this

Time-based detection of changes to multivariate patterns. / Hu, Jing; Runger, George.

In: Annals of Operations Research, Vol. 174, No. 1, 02.2010, p. 67-81.

Research output: Contribution to journalArticle

@article{27ebe9ca640446d48c1c6dee1fe64113,
title = "Time-based detection of changes to multivariate patterns",
abstract = "Detection of changes to multivariate patterns is an important topic in a number of different domains. Modern data sets often include categorical and numerical data and potentially complex in-control regions. Given a flexible, robust decision rule for this environment that signals based on an individual observation vector, an important issue is how to extend the rule to incorporate time-based information. A decision rule can be learned to detect shifts through artificial data that transforms the problem to one of supervised learning. Then class probability ratios are derived from a relationship to likelihood ratios to form the basis for time-weighted updates of the monitoring scheme.",
keywords = "Multivariate patterns, Supervised learning, Time-based detection of changes",
author = "Jing Hu and George Runger",
year = "2010",
month = "2",
doi = "10.1007/s10479-009-0610-8",
language = "English (US)",
volume = "174",
pages = "67--81",
journal = "Annals of Operations Research",
issn = "0254-5330",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Time-based detection of changes to multivariate patterns

AU - Hu, Jing

AU - Runger, George

PY - 2010/2

Y1 - 2010/2

N2 - Detection of changes to multivariate patterns is an important topic in a number of different domains. Modern data sets often include categorical and numerical data and potentially complex in-control regions. Given a flexible, robust decision rule for this environment that signals based on an individual observation vector, an important issue is how to extend the rule to incorporate time-based information. A decision rule can be learned to detect shifts through artificial data that transforms the problem to one of supervised learning. Then class probability ratios are derived from a relationship to likelihood ratios to form the basis for time-weighted updates of the monitoring scheme.

AB - Detection of changes to multivariate patterns is an important topic in a number of different domains. Modern data sets often include categorical and numerical data and potentially complex in-control regions. Given a flexible, robust decision rule for this environment that signals based on an individual observation vector, an important issue is how to extend the rule to incorporate time-based information. A decision rule can be learned to detect shifts through artificial data that transforms the problem to one of supervised learning. Then class probability ratios are derived from a relationship to likelihood ratios to form the basis for time-weighted updates of the monitoring scheme.

KW - Multivariate patterns

KW - Supervised learning

KW - Time-based detection of changes

UR - http://www.scopus.com/inward/record.url?scp=76649095286&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=76649095286&partnerID=8YFLogxK

U2 - 10.1007/s10479-009-0610-8

DO - 10.1007/s10479-009-0610-8

M3 - Article

AN - SCOPUS:76649095286

VL - 174

SP - 67

EP - 81

JO - Annals of Operations Research

JF - Annals of Operations Research

SN - 0254-5330

IS - 1

ER -