It is well known that different types of crime will have different impacts on different groups of people. For example, some groups of people will have higher risks of becoming victims of street crime simply because they spend more time in places where they come into contact with potential offenders. However, the locations of vulnerable groups of people vary considerably in both space and time.
Therefore, an accurate estimate the population at risk of crime in a given place and time is vital for a reliable estimate of the rate of crime. I.e., given that there were 10 crimes in a place at a particular time, we need to know how many potential victims there were to know whether the rate of crimes per person is high or (relatively speaking) low. However, estimating the number of potential victims is fraught with difficulty because data describing the movements of people, rather than simply where people live (like the census), are limited.
This research makes use of new ‘crowd-sourced’ data in an attempt to create more accurate estimates of the population at risk of mobile crimes such as street robbery. Importantly, these data are both spatially and temporally referenced and can therefore be used to estimate a crime rate in both space and time. Spatio-temporal cluster hunting techniques will be used to identify crime hotspots that are significant given the size of the ambient population in the area at the time.
The full paper is available in the journal Crime Science (here) and is free to access.
** Keywords**: Crime analysis and mapping; Population at risk; Clustering; Big data; Twitter; SatScan