top of page

Crime Map Analyst (CMA)

A two year project for the Home Office entailed the development of a toolkit (entitled Crime Map Analyst) that enabled non-GIS experts to undertake sophisticated spatial and temporal analysis. Functionality included:

Back to Paul's Research page

  • Density surfaces (hot and cold spots)

Hot and cold spots using kernel density estimation (KDE) without the requirement for complex parameters. Users were simply able to draw a rectangle over the study area

  • Repeat victimization identification

For finding spatial repeats based on geo-coordinates (such as the same building) or textual repeats (the same person or vehicle). It is also possible to constrain repeats to those that have occurred within a specified time window of previous events (for example within six months of each other)

  • Temporal analysis

Temporal analysis could be performed for months, hours, day of the week and so forth and produce graphical representations

  • Area profiles

The production of summary profiles  that allowed comparison of data at various spatial scales

  • Original / destination analysis

To explore the travel to and from crime. For example, where cars were being stolen and subsequently abandoned.

  • Ancillary functions

Including metadata editor (to maintain key characteristics of your data), raster to polygon conversion tools, advanced textual searches (to find similar crimes such as based on the modus operandi) and a detailed help system.

Some example output from the CMA toolkit can be found below:

​

GIS KDE density mapping generated by Paul Brindley
Temporal pattern of car crime - by Paul Brindley
Temporal patterns of crime (car crime) - by Paul Brindley

Figure 1: Dynamic zooming capability where users can draw a rectangle to zoom in to explore the spatial patterns in more detail

The changing geography of crime (domestic burglary) in Sheffield (by Paul Brindley)

Domestic burglary in 1998                                      Domestic burglary in 2000

Figure 2: Change in domestic burglary, 1998-2000. The red areas are increasing whilst the blue areas have decreasing levels of burglary. Such an approach allows for an effective visual tool for exploring trends over time.

Figure 3: Hourly trends of car crime in South Yorkshire, 1998 - 2000

Figure 4: Monthly trends of car crime in South Yorkshire, 1998 - 2000

bottom of page