« Back to News

Shotspotter: An Amazing Big Data Use Case To Tackle Gun Crime

May 23, 2016



One of the most interesting and positive uses I have seen of Big Data and analytics technology is ShotSpotter. The technology, developed by US-based SST Inc., works by effectively analyzing the entire soundscape of a city, and providing real-time alerts when gunfire is detected.

Things have moved quickly for ShotSpotter – in two years their coverage has grown from 30 cities in the US to 90 around the world. Crucially, the company has just announced a partnership with GE which will see the technology installed into the industrial giant’s Intelligent LED Smart City street lights, a key element of GE’s vision for modern, connected, Internet of Things-driven urban development.

But this is just one potential use for the technology – as well as cutting down on gun crime in the US, the company has high hopes that it will prove useful for countering terrorism in Europe and the Middle East, poaching in Africa and even help save coral reefs in south east Asia.

I spoke to president and CEO Ralph Clark, who told me that the system’s success has thrown up some insights which are changing the way police forces think about tackling gun violence.

“Two things I am really proud of”, he told me – “One of them was just really confirming something we already knew – which is that gunfire is a significantly underreported phenomena that happens a lot more frequently than people expect.”

“The other thing has been observing the decreases that agencies have experienced tackling gunfire using our system – we’re sophisticated enough to know that isn’t all just down to the technology – it’s about how agencies have deployed it as part of their wider gun violence abatement strategies – and in cities that we’ve analysed we’ve seen immediate reductions [in gun crimes] of 28%”. And perhaps the most interesting insight is that, by collating statistics from across multiple areas where ShotSpotter is installed, analysts have been able to infer that, while there may be a far higher number of gunshot incidents than was previously thought, there may actually be less shooters.

“What we’ve found by talking to these agencies is that it is very common to find that it is very few shooters carrying out a vast majority of these shootings. That’s very, very powerful – we know we don’t have to convince 4,000 people to shoot their guns less, we only have to deter maybe 10 or 15 people. By deter I mean either capture and convicting them, or letting them see that their friends are being captured and convicted. They are usually smart enough to figure out that they won’t get away with pulling the trigger as frequently.”

Clark is obviously aware that the value of ShotSpotter as it is used today is as much as a deterrent as it is a tool for catching gun toting criminals in the act.

The system is now in action across urban zones where gun violence is known to be a problem as diverse as Milwaukee and Puerto Rico. It works using sound sensors (Clark prefers this term to “microphones”, for reasons I shall get to …) positioned at strategic locations in affected neighborhoods. When a soundwave which matches the profile of gunfire is detected by three sensors, then its precise location can be ascertained by measuring the difference in time it takes for the sound to reach each sensor. This analysis is reported to the ShotSpotter control facility, where a final layer of human verification is applied before a report is raised with law enforcement.

Another benefit this brings, Clark tells me, is that it increases the visibility of police in areas where they often do not have a great reputation – a problem often brought about by their lack of visibility.

“Look at the fire service”, he explains to me, “There’s often a lot of respect for them, even in underserved communities. And that’s because they show up.

“It creates a bad feeling between communities and police when there is trouble and they don’t show up. If our tech is used in the right way it can really be an equalizer of service – they can provide the same level of service to underserved communities with gun violence, as they serve to more affluent communities with, say, property crime.”

Beyond this, the GE deal has the potential to expose ShotSpotter to a new range of use cases. GE will roll out its new smart, intelligent street lighting system – designed to conserve energy by operating smartly, for example switching off when no one is around – in smart city pilot projects around the world.

“Most of the technology was already in the lights”, Clark says, “GPS, the analogue to digital convertor, pretty much everything we needed to run ShotSpotter apart from the microphone.”

Rather than a manned service model, which ShotSpotter has seen such success with in recent years, this will see them switching to providing a software-and-support-only service. “We can think about pricing in a very different way. With GE we can afford to price on a per-city basis [the service is currently priced per square mile of coverage] which opens up new markets for us.

European cities, for example, where the gun violence problem is nothing like the scale of the US, would find applications in counter terrorism. This would mean that very different types of environments would be monitored than the economically deprived, residential urban areas which are generally monitored in US cities. “We’re very excited about this partnership and new opportunities to serve cities and create more safe environments for people who live in cities,” says Clark.

To further prove its adaptability to environments other than the ones for which it was originally designed, the system has also recently been successfully put to use across the vast subtropical plains of South Africa’s Kruger National Park. It has already been credited with facilitating the capture and prosecution of poachers targeting the area’s highly endangered rhino population.

And at this time company co-founder Robert Showen is in south east Asia where the system (in a modified form using technology similar to sonar) is being put to use fighting the scourge of blast fishing, which is causing irreparable damage to coral reefs. “When the reef goes, you see precipitous declines in the amount of fish available and that’s a protein source that a lot of people depend on – so we’re dealing with a real food security issue. So we are hoping that we will be a part of the solution to this awful issue.”

When ShotSpotter first came to my attention a few years ago, I was immediately impressed at the ambition and potential for driving positive change of this application of data analytics. It’s great to see that things have progressed and the company is looking for new challenges where its work could be helpful. Of course I had to ask Clark to address the privacy issues raised by his technology. After all, on the face of it, installing microphones to record every sound taking place in a city is surveillance on a rather grand scale.

He admits that there is a group of people who have raised concerns. “They call them ‘microphones’, we call them ‘sensors’”, he tells me. “There is a difference. Our sensor are purpose built to ignore ambient noise. When you are trying to detect specific noise, ambient noise is really the enemy. We have a very, very narrow use case for our surveillance. We say, ‘if you guys are so worried about surveillance, you should really be more worried about the microphone that you carry around in your pocket’. And then we refer them to our privacy policy – we think we’re in the right place.”

Forbes Magazine