Cross-correlation measure for mining spatio-temporal patterns

James Ma, Dajun Zeng, Huimin Zhao, Chunyang Liu

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Spatio-temporal data mining is finding applications in many domains, such as public health, public safety, financial fraud detection, transportation, and product lifecycle management. Correlation analysis is an important spatio-temporal mining technique for unveiling spatial and temporal relationships among multiple event types. This paper presents a new measure for assessing and analyzing spatio-temporal cross-correlations. This measure extends Ripley's K(r), a widely used measure of spatial correlation, with an additional temporal dimension. Empirical studies using real-world data show that the new measure can lead to a more discriminating and flexible spatio-temporal data analysis framework. In contrast with its predecessor, this measure also allows the discovery of leading (and potentially causal) event types whose occurrences precede those of other event types. Findings from analyses employing this measure may bear important managerial implications.

Original languageEnglish (US)
Pages (from-to)13-34
Number of pages22
JournalJournal of Database Management
Volume24
Issue number2
DOIs
StatePublished - Apr 2013

Fingerprint

Public health
Data mining

Keywords

  • Cross-correlation
  • Data mining
  • Public health
  • Public safety
  • Spatio-temporal data analysis

ASJC Scopus subject areas

  • Information Systems
  • Hardware and Architecture
  • Software

Cite this

Cross-correlation measure for mining spatio-temporal patterns. / Ma, James; Zeng, Dajun; Zhao, Huimin; Liu, Chunyang.

In: Journal of Database Management, Vol. 24, No. 2, 04.2013, p. 13-34.

Research output: Contribution to journalArticle

Ma, James ; Zeng, Dajun ; Zhao, Huimin ; Liu, Chunyang. / Cross-correlation measure for mining spatio-temporal patterns. In: Journal of Database Management. 2013 ; Vol. 24, No. 2. pp. 13-34.
@article{fab01f3aa3fe48ecad04f5e465216d9a,
title = "Cross-correlation measure for mining spatio-temporal patterns",
abstract = "Spatio-temporal data mining is finding applications in many domains, such as public health, public safety, financial fraud detection, transportation, and product lifecycle management. Correlation analysis is an important spatio-temporal mining technique for unveiling spatial and temporal relationships among multiple event types. This paper presents a new measure for assessing and analyzing spatio-temporal cross-correlations. This measure extends Ripley's K(r), a widely used measure of spatial correlation, with an additional temporal dimension. Empirical studies using real-world data show that the new measure can lead to a more discriminating and flexible spatio-temporal data analysis framework. In contrast with its predecessor, this measure also allows the discovery of leading (and potentially causal) event types whose occurrences precede those of other event types. Findings from analyses employing this measure may bear important managerial implications.",
keywords = "Cross-correlation, Data mining, Public health, Public safety, Spatio-temporal data analysis",
author = "James Ma and Dajun Zeng and Huimin Zhao and Chunyang Liu",
year = "2013",
month = "4",
doi = "10.4018/jdm.2013040102",
language = "English (US)",
volume = "24",
pages = "13--34",
journal = "Journal of Database Management",
issn = "1063-8016",
publisher = "IGI Publishing",
number = "2",

}

TY - JOUR

T1 - Cross-correlation measure for mining spatio-temporal patterns

AU - Ma, James

AU - Zeng, Dajun

AU - Zhao, Huimin

AU - Liu, Chunyang

PY - 2013/4

Y1 - 2013/4

N2 - Spatio-temporal data mining is finding applications in many domains, such as public health, public safety, financial fraud detection, transportation, and product lifecycle management. Correlation analysis is an important spatio-temporal mining technique for unveiling spatial and temporal relationships among multiple event types. This paper presents a new measure for assessing and analyzing spatio-temporal cross-correlations. This measure extends Ripley's K(r), a widely used measure of spatial correlation, with an additional temporal dimension. Empirical studies using real-world data show that the new measure can lead to a more discriminating and flexible spatio-temporal data analysis framework. In contrast with its predecessor, this measure also allows the discovery of leading (and potentially causal) event types whose occurrences precede those of other event types. Findings from analyses employing this measure may bear important managerial implications.

AB - Spatio-temporal data mining is finding applications in many domains, such as public health, public safety, financial fraud detection, transportation, and product lifecycle management. Correlation analysis is an important spatio-temporal mining technique for unveiling spatial and temporal relationships among multiple event types. This paper presents a new measure for assessing and analyzing spatio-temporal cross-correlations. This measure extends Ripley's K(r), a widely used measure of spatial correlation, with an additional temporal dimension. Empirical studies using real-world data show that the new measure can lead to a more discriminating and flexible spatio-temporal data analysis framework. In contrast with its predecessor, this measure also allows the discovery of leading (and potentially causal) event types whose occurrences precede those of other event types. Findings from analyses employing this measure may bear important managerial implications.

KW - Cross-correlation

KW - Data mining

KW - Public health

KW - Public safety

KW - Spatio-temporal data analysis

UR - http://www.scopus.com/inward/record.url?scp=84891759501&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84891759501&partnerID=8YFLogxK

U2 - 10.4018/jdm.2013040102

DO - 10.4018/jdm.2013040102

M3 - Article

AN - SCOPUS:84891759501

VL - 24

SP - 13

EP - 34

JO - Journal of Database Management

JF - Journal of Database Management

SN - 1063-8016

IS - 2

ER -