Crime-stopping pays for Contré

greg

It reads like a Hollywood screenplay.

After joining the military at 17 and traveling the globe, soft-spoken and mild mannered Stephane Contré from Quebec City becomes a beat cop in Ottawa. His hobby is tinkering with technology and caressing computer code to help him to do his job better. One day, it will enable authorities to fight crime in a way no one really thought possible: to predict when and where it would happen… and to stop it before it occurs.

After three years in our nation’s capital, Contré finds himself deep in North Central Africa, in the Republic of Chad. He is many time zones away from Canada’s House of Parliament and even farther away from his wife Tia. She has returned to her hometown, Edmonton, where the couple met when he was posted at Griesbach with the Military Airborne School.

In Chad, Contré is a security advisor for EnCana Corp. on an oil and gas exploration project, mitigating security issues. He recalls, “This is where things started to percolate… looking to see where and when things might occur. I was looking for more attributes within the criminal space that would lead to better forecasting and allow us to better manage our security forces.”

Two years later—and before things really get off the ground—he faces another challenge: His position abruptly comes to an end.

Perhaps it is a blessing in disguise. Contré is over 11,000km away from his wife… it’s “a 32-hour flight”… they see each other every 35 days.

IDENTIFYING CRIME HOT SPOTS

Being home “also gave me the time I needed to work on my models.”

And work he did: evenings and weekends and any available time. Tia recalls one stretch when she barely saw him from before Christmas until the first week of January.

Unfazed, she was fully supportive. “I’ve always known him to be this way. If he wasn’t doing what he’s doing, I’d wonder what was wrong with him. At least, I know he’s home and not at the bar.”

But Tia didn’t just sit around and watch him work. She partnered with him.

“She’s my sounding board,” says Contré. “Furthermore, I’m a cop first and a programmer second.” And, she’s better at handling the business affairs. “My wife is from a family of entrepreneurs.”

“I encourage him and rein him in,” she says. He adds, “Like a manager and a boxer—you know, like Rocky.”

In 2005, Contré joined Edmonton Transit System (ETS) as a crime intelligence analyst. “It was a perfect place to pilot this project”—one he had been working on for years. “I had an opportunity to validate my models in a live environment.”

His vision became the Daily Crime Forecast (DCF), Windows-based software designed to assist police and security agencies in deployment of patrol resources.

Contré says that agencies—depending on their method of managing intelligence—typically will deploy to specific locations, commonly known as ‘crime hot-spots’ because there are chances of a crime occurring. These exist because spatially there is a clustering of incidents at the same place. Contré takes this concept and pushes it to another dimension.

“If you look past the crime data—data collected by an agency—you can look spatially where things happen, but also temporally.” Temporal factors or attributes include time of day, day of the week, day of the month, month of the year within incident data.

“All of these things are relevant in determining when a crime is likely to occur.” And, no, factors such as weather and a full moon are not statistically significant, according to Contré’s data sets.

By applying proprietary algorithms and performing data mining procedures, a process of extracting hidden patterns from data, DCF produces a threat map or crime-mapping component. “It tells the officer where and when to deploy to maximize efficiency. (Asking) ‘where can I go right now to be sure or be most effective in terms of deterring crime or apprehending criminals?’”

During a special event, like a rock concert, resources will naturally be deployed. But, the forecast becomes invaluable when it’s just a regular day. “Like a Saturday in a certain area in April. If you look back, you can see definite trends occurring and capitalize on it.”

Take, for example, Whyte Avenue: “It gets ‘hot’ in the evening, and when things taper down, the ‘hot-spot’ moves toward the university… drunks going home and kicking garbage cans. Agencies then need to see where [they’ll get] the best bang for their buck and that’s where their officers go.”

Contré calls it using proactive deployment times more efficiently—in essence, arriving before the actual event and preventing the crime from occurring.

This underlying theme is slightly reminiscent of Chief John Anderton of the pre-crime squad in Washington DC. They too used advanced technology and supernatural beings known as “precognitives” to prevent crime. Of course, that was a Spielberg movie: The Minority Report was set in 2054 and Anderton was played by Tom Cruise.

While The Minority Report is based on a fictional short story, Contré’s crime forecast is real and is being used today. To keep the information fresh, the forecast for the next 48-hour period is regenerated every day between 4 and 5AM when crime is typically at a lull, taking into account any new data introduced in the last 24hours.

Contré tells us that the DCF has been deployed to a live environment within the Edmonton Transit System where peace officers have willingly adopted the model to guide their patrols. Both managers and supervisors also use it to readily identify resource allocation priorities within the transit system.

“The software gives me reassurance that my resources are deployed in the most efficient and effective fashion possible,” says Ron Gabruck, director of ETS safety and security.

If a patrol is present, it’s less likely an issue will occur and, if it does, it can be dealt with before it becomes a larger issue. Gabruck calls this the “broken window” theory. “We deal with—and spend a lot of out time on—a lot of minor nuisance type crimes such as drinking in public, swearing and horseplay.” But ETS peace officers get there immediately to clean it up.

“If you allow this group of individuals to be in a station, then allow them to sit there and drink… to get drunk, the path could lead to crime—assaults, mischief and perhaps even robberies.”

Gabruck stresses the positive impact of being there early, impressing that perception of the safety in the system—especially in buses and transit centres—is relatively high; they strive to keep it that way.

“I have extreme confidence that my resources are where they should be when they should be. As a manager in these fiscal times, I’m please to be able to say others (transit systems) are envious.”

Gabruck acknowledges it’s a vast network and the stats speak for themselves… that, relatively speaking, crime is down. While he admits this could be from a number of factors, he is convinced that the DCF model is contributing to the reduction in crime. It “is just one part of the ETS response system… a great complement to the overall system.”

Contré’s software has changed the way crime is handled. Proactive calls (where officer already has been deployed) are up 159 percent; reactive calls (where people call in to complain) are down 52 percent since implementation. The pre-emptive arrival of officers has clearly resulted in fewer people calling in after the fact.

Gabruck speaks highly of Contré and his creation, referring to him as “a genius… an incredibly talented individual. We’re lucky to have him.”

AWARD-WINNING RECOGNITION

For Contré, stopping crime definitely has its rewards. Use of the daily crime forecast within ETS won the Canadian Urban Transit Association Award for Innovation. And, last December, Contré and his DCF took first place in the inaugural novaNAIT Technology Commercialization Challenge The grand prize: $10,000 in novaNAIT services, including business development assistance and mentoring.

novaNAIT is about industry driven applied research,” says Dr. Sam Shaw, NAIT’s president and CEO. “Stephane Contré’s software fits perfectly with this model; as the winner of the competition, we are going to help him commercialize the technology he has developed.”

Finding out about the contest was a bit of a fluke. Contré explains, “I happened to be living in the right place at the right time.” Living being the operative word—NAIT’s TechLife magazine was still being delivered for the previous occupant who hadn’t left a forwarding address.

When he noticed a picture of Dr. Shaw inviting people with a good idea to compete in the challenge, he decided to enter. Within weeks, Contré received an e-mail indicting he was one of eight finalists.

“We had to present to a board of technology and business experts—five minutes to present and five for questions. I was able to demo the software.”

Once again the story played out to a Hollywood end: 24 hours later, the director left a message on the answering machine. Tia delivered it to Contré, who was in the shower. “You won!” she said, with a huge grin on her face.

“Holy cow! Ten Grand!”

Contré admits that he loves his job at ETS. He’s continually looking to find ways to make the DCF even better, hoping that other agencies, like the police service, will look at using DCF. He’s also investigating the possibilities of partnering or licensing to another company, maybe one in the police and security area.

Sort of like the scenario played out in the CBS drama NUMB3RS: Charlie Eppes, a brilliant math professor uses the science of mathematics with its complex equations to help his brother, Don, an FBI special agent, ferret out criminals and solve baffling crimes. Is there a comparison to what Contré does?

“I’ve been told that before… to some extent, but that guy is waaaaaaaay smarter than I ever will be!”

Except that Charlie Eppes is pretend smart—Stephane Contré is real smart. √

Greg Gazin, “The Gadget Guy”, is a serial entrepreneur, freelance technology columnist, small business speaker, an avid Podcaster and producer of Toastcaster.com. Greg can be reached at 780.424.1881, gadgetgreg.com or greg@gadgetguy.ca

For further information, contact
Stephane Contré
780.432.5311
e-mail: Stephane.Contré@crimeforecast.com
http://www.crimeforecast.com/


%d bloggers like this: