About Us

Overview

Aim: introduce a number of methods to understand and model crime

Mixture of short lectures and practical sessions

A full description is impossible!

Briefly introduce methods and highlight where to go next

You can choose which methods are most suitable for further study

Standard Risk Surface Analysis: Assessing Crime

Tool: CrimeStat

A very useful tool for performing common spatio-temporal analyses

Hot-spot mapping (e.g. Kernel Density Estimation)

Geographic Profilling

Space-Time Analysis

Journey to Crime

.. mnany others ..

Excellent documentation

No practical session organised, but help is available

RTM: Adding the Environment

Tool: RTMDx

Risk Terrain Modelling: GIS techniques to explore the relationship between crime and spatial factors that influence it. E.g.

Public transport

Drug dealers

Schools

Gangs

Parks

Shops

...

Can be used to make predictions

Practical session: Generate a risk surface

Near-repeat modelling: adding behaviour

Method: Trafford Model

Near Repeats

~31% to 76% criminals return to commit further crimes

Crimes clustered in space and time

Criminals return to an area over a few weeks until risk rises or reward drops

Might be thought of as an "optimal forager" strategy

If we can predict their return pattern we can prevent the crimes

Johnson & Bowers (2004); Bernasco (2008)

Practical session: Near repeat calculator and near repeat mapping

Microsimulation: Adding Victims

Tool: Flexible Modelling Framework

Method for estimating populations at the individual level

Also allows us to estimate variables missing in an area

Practical session: modelling a city population

ABM: bringing people and the environment together

Tool: NetLogo

Agent-Based Modelling: A method to simulate the behaviour of individuals

Phenonema are 'grown' from the ground-up

Well suited to crime analysis

Example: Simulating burglary

Practical: Using NetLogo to build an agent-based model