Predicting Crime Patterns Using Big Data

Predicting Crime Patterns Using Big Data

Researchers have long been using previous crime data to study and understand the crime pattern and the behaviour of the perpetrators. This has formed the plot of several crime thrillers as well. But predicting future crime pattern using the already available data is something new and gaining much popularity with the cops, researchers and governments around the world.

Forecasting Crime Pattern

Can the occurrence of a crime be actually predicted using the earlier patterns? Trials by a few police departments and governments using computer programmes and big data are pointing towards such as possibility that could be exploited to reduce criminal activities.  As per the report of BBC iWonder, the Los Angeles and Manchester cops have run a trial test of such a crime prediction in 2011. They used a large amount of crime data to spot patterns in people’s behaviour that could ultimately help them spot the crime before it occurs. Further, they deployed police personnel at the place of prediction. According to the report, the trial was a success with a remarkable decline in crime and burglaries.

The crime pattern prediction is based on the principle of predictive analytics whereby huge amount of data and Big Data mining techniques such as statistics, modelling and machine learning are used to find patterns that throw light into the current or future human behaviour. The process first defines the average behaviour of a population and then matches individuals to that pattern.

Radical Trials

Predictive policing trials have taken place under various police departments in many countries such as UK, China and the US. In UK, many counties have experimented with the Big Data and computer algorithms to spot patterns in crime.

In an attempt to thwart anti-nationals and terrorists from seeding instability in the country, the Chinese government has sought the assistance of the state-run defense contractors, China Electronics Technology Group. As per the report published by Bloomberg, the technology giant is entrusted with the task of developing software to collect details on the jobs, consumption habits, hobbies and other behaviours of its citizens. It would then flag any sort of unusual behaviour such as a huge amount of money suddenly found in a poor citizen’s account or unusual frequent calls to a foreigner by someone who doesn’t have relatives or friends in foreign countries. They hope to use several such indicators to predict and prevent seditious activities, if any, before they actually happen.

PredPol, a predictive policing system developed by P Jeffrey Brantingham, a professor of anthropology at University of California, Los Angeles, is now being used by several police departments in the US and UK.  Los Angeles Police Department officials are using the prediction outcome from the software to contain high possibility criminal activities such as burglary, vehicle break-ins and car theft within their jurisdiction.

Is crime prediction a success?

While the trials on crime prediction using big data are gaining momentum in different parts of the world, there are mixed responses about its efficacy as well. The trials in UK largely point towards the success of the predictive policing model.  However, this seems to be effective where the criminals exhibit predictable patterns of behaviour. According to the BBC report, in Trafford, Manchester, a 26.6% decline in burglaries compared to a 9.8% fall across Greater Manchester was recorded in 2011. However, Kent Police had a different experience. The experimental run they had for four months from December 2012 marked an increase in crime for the following year. Erroneous crime data coupled with the failure to deploy the cops efficiently were said to be the reason for failure in the model.

Fighting cyber crimes

With notable increase in cyber-crimes, and security risks becoming bigger and bigger, it is all the more important for enterprises to stay safe and secure today. The big data analytics with its ability to accurately analyse threats and incidents from different sources help companies to combat cyber-crimes. There are many companies that make use of big data analysis for aiding their business processes.

Ethical Issues

While many see huge potential for predictive policing, there are those who raise concerns about the invasion of privacy. They argue that once it becomes a valid law enforcement tool, authorities will be able to collect any sort of sensitive data concerning a person, which will be an invasion into their privacy. Although the data used currently is anonymous, it can actually be misused to track specific people.

If you would like to know more about big data and its usage in the prediction of criminal activities, send e-mail to

Author : Letterbug Date : 02 Jun 2016