Following last week’s horrible attacks in Barcelona, a friend asked me what IoT technology is doing to fight violent crime. While I was hardly surprised by the question, I had to admit that I didn’t know the answer. It got me thinking about the potential of big data analytics to spot patterns and trends and how this information could be used to identify when and where violent crime is most likely to occur. With enough data, and the means to interpret it, it might even be possible to prevent crime before it happens.
The good news is that police departments are beginning to use big data, machine learning and predictive analytics to understand and prevent crime giving them the opportunity to deploy police resources in response to anticipated threats.
Knowledge is power: insights from big data
We live in a world under constant surveillance. In many countries and across Europe CCTV is on every street corner, shopping mall, and liquor store. While this ubiquitous scrutiny may make citizens uneasy, it does have one big advantage: that those who commit crimes will probably be spotted in the act. But this does little to prevent the crime from happening in the first place.
In the U.S., several police-led initiatives are making the most of surveillance information. With IoT data police can analyze crime patterns and trends. By applying predictive analytics and machine learning to vast sets of data, police departments can more easily forecast where and when violent crime will break out, and ensure that they have the resources in place to prevent it.
The Chicago Police Department is applying machine learning and predictive analytics to police data sets; including crime incidents, arrests, and weather data. When historical data (like previous arrest records) is combined with real-time IoT data, such as sensor-influenced cameras designed to detect gunshots, it becomes easier to pinpoint problem locations and understand the conditions in which crime can flourish.
Bringing this information together is HunchLab: a geographic prediction tool that uses data modeling to predict risk in specific locations across the city. At-risk areas are highlighted on-screen, while recommendations for evasive action (such as deploying a high visibility police patrol car to take stock of the situation and deter criminals) are displayed alongside. This information is collated into a ‘decision support system’, made available to individual police officers on the beat.
IBM i2 Coplink: connecting police officers
One example of such a decision support system is Coplink, or to give it its full title: IBM i2 Coplink. This tool consolidates disparate data sets (such as arrest records, mugshots, location data and known gang affiliations) into a single dashboard, from which police across different locations can view and share vital information easily and securely. This reduces the risk of information slipping through the cracks in a complex investigation.
So far, this predictive approach has worked best against burglary and contents from parked cars. Manchester, for instance, reported reductions of 12% in robberies, 21% in burglaries, and 32% in theft from vehicles, following the adoption of recommendations for preventive action from statistical analysis.
Part of the reason for the skewed success towards common crimes like these is that they yield plenty of historical data, which can easily be supplemented with other information. Road network maps, for instance, show which areas are easily accessible and can offer a quick getaway, and which are more closed off.
Weather data, too, plays a big part in predicting when robberies will take place. Robbers don’t like rain, apparently, so fair weather days are more likely occasions for crime.
How proactive policing can prevent crime
The importance of tone: uncovering threats in social media
Of course, crimes that take place in the open are just one side of the coin. On the other side are those that are harder to anticipate – either because they appear to be random, isolated incidents, or because their perpetrators operate with the protection of larger organizations that can hide them from view.
Some attacks of this nature do share common characteristics, however, which can help flag them up in advance. One of these is the social media brag. Would-be attackers or terrorists who can’t resist showing off on Facebook leave valuable traces for those who would catch them: spot the brag, bag the terrorist, prevent the attack.
There are two challenges, however, to this approach. The first is the sheer volume of social media content. Post numbers run into the billions, and sifting through them all is no mean feat.
The second is that a lot of people say things they don’t mean on social media. Separated from face-to-face contact by a protective and sometimes anonymizing distance, it’s easier to post inflationary, anger-fueled content than it would be to say something of the sort in person. So how can we differentiate between what is dangerous, and what is merely unpleasant? Can technology help make that distinction?
The Tactical Institute, home to specialists in real-time predictive threat detection, may have the answer. The Institute’s job is to identify threats on social media and pre-empt threatening or violent incidents. Among its team of analysts are a number of combat-wounded veterans, who are trained to evaluate posts for criminal intent, and to establish whether the people behind them have the means and opportunity at their disposal to carry it out.
Carrying out this work is a slow process. Watson Analytics for Social Media analyzes the social sentiment of user-generated content, identifying warning signs and flagging problem posts for further review. This means that only pertinent information is passed on to Tactical Institute staff, significantly reducing the time it takes to identify genuine threats.
It will never be possible to entirely eliminate threat or crime from our lives. But the IoT can go some way into offering vigilant, dependable tools that support police departments and other organizations as they do their best to keep us safe.
If you’re interested in learning more about these crime prevention tools, take a look at the resources below:
Welcome to Day 2 of the IBM IoT Exchange. Another day in which our four academies delivered valuable insights, great main stage presentations and demos that brought technology to life. And if you missed yesterday’s recap, you can read it here. Just like yesterday, we wanted to share some of the day’s highlights: Day 2 Keynote: ...read more
Welcome to Day 1 of the IBM® IoT Exchange. With four academies, main stage presentations and a robust solutions showcase, it was an information-packed day, with IoT thought leaders and experts, and IBM customers, IBM Business Partners and staff. Did we mention we had a backyard party, too? More on that later. If you’re here with ...read more
Over the last 20 years, huge investments have been made in metropolitan areas to understand the immense volume of data collected about the citizens inhabiting these cities. Hundreds of millions of dollars have been spent combining data with emerging mobile, sensor and machine learning technologies. This is all done with the goal to improve infrastructure ...read more