Exploring spatial patterns of crime using non-hierarchical cluster analysis

Alan T. Murray, Anthony Grubesic

Research output: Chapter in Book/Report/Conference proceedingChapter

9 Citations (Scopus)

Abstract

Exploratory spatial data analysis (ESDA) is a useful approach for detecting patterns of criminal activity. ESDA includes a number of quantitative techniques and statistical methods that are helpful for identifying significant clusters of crime, commonly referred to as hot spots. Perhaps the most popular hot spot detection methods, both in research and practice, are based on tests of spatial autocorrelation and kernel density. Non-hierarchical clustering methods, such as k-means, are less used in many contexts. There is a perception that these approaches are less definitive. This chapter reviews non-hierarchical cluster analysis for crime hot spot detection. We detail alternative non-hierarchical approaches for spatial clustering that can incorporate both event attributes and neighborhood characteristics (i.e., spatial lag) as a modeling parameter. Analysis of violent crime in the city of Lima, Ohio is presented to illustrate this for hot spot detection. We conclude with a discussion of practical considerations in identifying hot spots.

Original languageEnglish (US)
Title of host publicationCrime Modeling and Mapping Using Geospatial Technologies
PublisherSpringer Netherlands
Pages105-124
Number of pages20
ISBN (Electronic)9789400749979
ISBN (Print)9789400749962
DOIs
StatePublished - Jan 1 2013
Externally publishedYes

Fingerprint

Crime
Cluster analysis
crime
cluster analysis
spatial data
detection method
Autocorrelation
autocorrelation
Statistical methods
modeling
method
data analysis
detection

Keywords

  • Clustering
  • Hot spots
  • Spatial patterns

ASJC Scopus subject areas

  • Earth and Planetary Sciences(all)
  • Engineering(all)

Cite this

Murray, A. T., & Grubesic, A. (2013). Exploring spatial patterns of crime using non-hierarchical cluster analysis. In Crime Modeling and Mapping Using Geospatial Technologies (pp. 105-124). Springer Netherlands. https://doi.org/10.1007/978-94-007-4997-9_5

Exploring spatial patterns of crime using non-hierarchical cluster analysis. / Murray, Alan T.; Grubesic, Anthony.

Crime Modeling and Mapping Using Geospatial Technologies. Springer Netherlands, 2013. p. 105-124.

Research output: Chapter in Book/Report/Conference proceedingChapter

Murray, AT & Grubesic, A 2013, Exploring spatial patterns of crime using non-hierarchical cluster analysis. in Crime Modeling and Mapping Using Geospatial Technologies. Springer Netherlands, pp. 105-124. https://doi.org/10.1007/978-94-007-4997-9_5
Murray AT, Grubesic A. Exploring spatial patterns of crime using non-hierarchical cluster analysis. In Crime Modeling and Mapping Using Geospatial Technologies. Springer Netherlands. 2013. p. 105-124 https://doi.org/10.1007/978-94-007-4997-9_5
Murray, Alan T. ; Grubesic, Anthony. / Exploring spatial patterns of crime using non-hierarchical cluster analysis. Crime Modeling and Mapping Using Geospatial Technologies. Springer Netherlands, 2013. pp. 105-124
@inbook{1545b89158fe4055ab91a07c22f62720,
title = "Exploring spatial patterns of crime using non-hierarchical cluster analysis",
abstract = "Exploratory spatial data analysis (ESDA) is a useful approach for detecting patterns of criminal activity. ESDA includes a number of quantitative techniques and statistical methods that are helpful for identifying significant clusters of crime, commonly referred to as hot spots. Perhaps the most popular hot spot detection methods, both in research and practice, are based on tests of spatial autocorrelation and kernel density. Non-hierarchical clustering methods, such as k-means, are less used in many contexts. There is a perception that these approaches are less definitive. This chapter reviews non-hierarchical cluster analysis for crime hot spot detection. We detail alternative non-hierarchical approaches for spatial clustering that can incorporate both event attributes and neighborhood characteristics (i.e., spatial lag) as a modeling parameter. Analysis of violent crime in the city of Lima, Ohio is presented to illustrate this for hot spot detection. We conclude with a discussion of practical considerations in identifying hot spots.",
keywords = "Clustering, Hot spots, Spatial patterns",
author = "Murray, {Alan T.} and Anthony Grubesic",
year = "2013",
month = "1",
day = "1",
doi = "10.1007/978-94-007-4997-9_5",
language = "English (US)",
isbn = "9789400749962",
pages = "105--124",
booktitle = "Crime Modeling and Mapping Using Geospatial Technologies",
publisher = "Springer Netherlands",

}

TY - CHAP

T1 - Exploring spatial patterns of crime using non-hierarchical cluster analysis

AU - Murray, Alan T.

AU - Grubesic, Anthony

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Exploratory spatial data analysis (ESDA) is a useful approach for detecting patterns of criminal activity. ESDA includes a number of quantitative techniques and statistical methods that are helpful for identifying significant clusters of crime, commonly referred to as hot spots. Perhaps the most popular hot spot detection methods, both in research and practice, are based on tests of spatial autocorrelation and kernel density. Non-hierarchical clustering methods, such as k-means, are less used in many contexts. There is a perception that these approaches are less definitive. This chapter reviews non-hierarchical cluster analysis for crime hot spot detection. We detail alternative non-hierarchical approaches for spatial clustering that can incorporate both event attributes and neighborhood characteristics (i.e., spatial lag) as a modeling parameter. Analysis of violent crime in the city of Lima, Ohio is presented to illustrate this for hot spot detection. We conclude with a discussion of practical considerations in identifying hot spots.

AB - Exploratory spatial data analysis (ESDA) is a useful approach for detecting patterns of criminal activity. ESDA includes a number of quantitative techniques and statistical methods that are helpful for identifying significant clusters of crime, commonly referred to as hot spots. Perhaps the most popular hot spot detection methods, both in research and practice, are based on tests of spatial autocorrelation and kernel density. Non-hierarchical clustering methods, such as k-means, are less used in many contexts. There is a perception that these approaches are less definitive. This chapter reviews non-hierarchical cluster analysis for crime hot spot detection. We detail alternative non-hierarchical approaches for spatial clustering that can incorporate both event attributes and neighborhood characteristics (i.e., spatial lag) as a modeling parameter. Analysis of violent crime in the city of Lima, Ohio is presented to illustrate this for hot spot detection. We conclude with a discussion of practical considerations in identifying hot spots.

KW - Clustering

KW - Hot spots

KW - Spatial patterns

UR - http://www.scopus.com/inward/record.url?scp=84935126996&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84935126996&partnerID=8YFLogxK

U2 - 10.1007/978-94-007-4997-9_5

DO - 10.1007/978-94-007-4997-9_5

M3 - Chapter

AN - SCOPUS:84935126996

SN - 9789400749962

SP - 105

EP - 124

BT - Crime Modeling and Mapping Using Geospatial Technologies

PB - Springer Netherlands

ER -