Security Journal

FIGURES AND TABLES

FROM:

The Utility of Hotspot Mapping for Predicting Spatial Patterns of Crime

Spencer Chainey, Lisa Tompson and Sebastian Uhlig

BACK TO ARTICLE
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Figure 1.

Common hotspot mapping techniques. (a) Point mapping, (b) standard deviational spatial ellipses, (c) thematic mapping of administrative units, (d) grid thematic mapping and (e) KDE.

Full figure and legend (148K)
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Figure 2.

The Camden and Islington study area in central/north London.

Full figure and legend (102K)
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Figure 3.

Hotspots were determined by selecting the uppermost thematic class calculated using the five classes and the default values generated from applying the quantile thematic range method in MapInfo.

Full figure and legend (55K)
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Figure 4.

Hotspot maps generated from 3 months of residential burglary input data (measurement date of the 1 January 2003) using (a) STAC, (b) thematic mapping of output areas, (c) grid thematic mapping and (d) KDE. Each map is shown with its PAI value, based on 1 month of measurement data.

Full figure and legend (127K)
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Figure 5.

KDE hotspot maps of (a) residential burglary and (b) street crime, generated from 3 months input data, and where the hotspot area in each is controlled to represent 3 per cent of the total area. Each figure is presented with its PAI value and its hotspot hit rate for predicting where crimes in the next month occurred. The street crime hotspot map is over twice as a good as the residential burglary hotspot map for predicting where crimes of the respective crime type may occur in the future.

Full figure and legend (48K)
Table 1 - Number of crime events, by type, for each calendar year - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Table 2 - Temporal slices of input data for generating hotspot maps, for (a) a measurement date of the 1 January 2003 and (b) a measurement date of the 13 March 2003 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Table 3 - Temporal slices of measurement data for calculating the ability of hotspot maps to predict spatial patterns of crime, for (a) a measurement date of the 1st January 2003 and (b) a measurement date of the 13 March 2003 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Table 4 - Hotspot Detective KDE default values (C.S. - cell size and B. - bandwidth) for each crime type and each period of input data, for (a) a measurement date of the 1 January 2003 and (b) a measurement date of the 13 March 2003 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Table 5 - PAI values for residential burglary, street crime, theft from vehicles and theft of vehicles - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Table 6 - PAI values for different hotspot mapping techniques - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Table 7 - PAI values for different hotspot mapping techniques, by crime type - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Table 8 - PAI and actual crimes predicted using kernel density estimation to generate a hotspot map from the previous three months of crime data and determine where crimes in the next month may occur (using a measurement date of the 1 January 2003) - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
BACK TO ARTICLE