Big ‘Mobile Phone’ Data predicts crimes before they happen
While it’s true that crime can occur anywhere and criminals come in all shapes and sizes, it’s still possible to find patterns in the data. Thus, while it can be difficult to predict if a specific individual is likely to commit a crime, attempts to determine when are where crimes might take place can be surprisingly accurate.
Predictive analytics is the data scientist’s answer to the crystal ball. Rather than staring into a glass ball and hoping to glean insights from some unknown force, what they do is peer into their data and use techniques like machine learning, modeling, and statistics to look for patterns that might indicate future behavior.
The science has become popular in recent years and often produces some impressive insights, such as the retailer Target being able to guess when its customers are pregnant before they even know themselves. However we should note this isn’t a perfect science – Google’s Flu Trends is a good example of this.
One of the more interesting use cases of predictive analytics is trying to predict future crimes. A recent paper, Once Upon a Crime: Towards Crime Prediction from Demographics and Mobile Data, authored by Andrey Bogomolov, Bruno Lepri, Jacopo Staiano, Nuria Oliver, Fabio Pianesi, and Alex Pentland, explores the possibility of using mobile phone data to predict when and where future crimes might occur.
“The main contribution of the proposed approach lies in using aggregated and anonymized human behavioral data derived from mobile network activity to tackle the crime prediction problem. While previous research reports have used either background historical knowledge or offenders’ profiling, our findings support the hypothesis that aggregated human behavioral data captured from the mobile network infrastructure, in combination with basic demographic information, can be used to predict crime. In our experimental results with real crime data from London we obtain an accuracy of almost 70% when predicting whether a specific area in the city will be a crime hotspot or not. Moreover, we provide a discussion of the implications of our findings for data-driven crime analysis.”
The author’s methods aren’t too dissimilar from those depicted in the movie, The Minority Report, something which could be viewed as controversial. Even so, few will argue the data is certainly interesting. Here’s a look at the author’s predicted crime map of London, UK:
The London crime map was derived from anonymized and aggregated human behavioral data take from mobile network activity within the London area, and computed using Telefonica Digital’s Smart Steps.
Telefonica says Smart Steps is a product that “provides insights based on the behavior of crowds to help companies and public sector organizations make informed business decisions”.
The authors used this tool and combined it with historical crime statistics and mobile network data to identify crime ‘hotspots’ in London. But aside from revealing hotspots, it also provides insights into where criminal activity is likely to happen, and even where it’s currently taking place.
There are downsides to this method though: It’s often the case that anonymized data isn’t always truly anonymous, which makes it possible to track specifc individuals – and that could lead to the kinds of abuses seen in The Minority Report.
Nevertheless, the science of geolocating future crimes does hold lots of promise, and could one day become a very useful tool for law enforcement agencies around the world.
photo credit: Funky64 (www.lucarossato.com) via photopin cc
A message from John Furrier, co-founder of SiliconANGLE:
Show your support for our mission by joining our Cube Club and Cube Event Community of experts. Join the community that includes Amazon Web Services and Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger and many more luminaries and experts.
We really want to hear from you, and we’re looking forward to seeing you at the event and in theCUBE Club.