From the New Scientist:
THE patrol car comes to a stop in a sleepy neighbourhood of small, earth-coloured homes. A woman saunters past pushing a stroller. It is daytime in the artsy beach town of Santa Cruz but I am still a little spooked. Futuristic crime prediction software sent me here with my companion, deputy chief of police Steve Clark. And it’s just possible that the tranquillity is due to our presence.
Rather than predicting who will commit crimes, like the fictitious “precrime” system from the 2002 film Minority Report, the software that the Santa Cruz police department has recently started field-testing looks at where crime might be committed.
It uses the locations of past incidents to flag up likely future crime scenes. Police can then target their patrols on these areas, in the hope that their presence might stop the predicted crimes from happening at all. At the very least, they will be on the spot to help victims and make arrests.
The program has been built by mathematician George Mohler, at Santa Clara University in California, and his colleagues. They noted that some crimes follow potentially predictable patterns. One burglary, for example, tends to trigger others nearby in the next few days, rather like aftershocks from an earthquake (see graph). In 2010, Mohler’s team turned equations used to predict aftershocks into the basis for a program that uses the dates and times of reported crimes to predict when and where the “aftercrimes” will occur.
On average the program predicted the location and time of 25 per cent of actual burglaries that occurred on any particular day in an area of Los Angeles in 2004 and 2005, using just the data on burglaries that had occurred before that day (Journal of the American Statistical Association, DOI: 10.1198/jasa.2011.ap09546).
Now the program is undergoing its first field test in Santa Cruz. Every day. it flags up 10 areas, each 150 metres square, for each of three types of crime – residential burglary, auto burglary and auto theft. Clark updates the program each night with new data and calculates the location probabilities for the next day.