The ping’s the thing
The Eagle County jail is home to two murder suspects whom police found by locating their cell phones.
Traci Cunningham was tracked to a coffee shop on the Front Range. She’s charged with murdering her adoptive mother.
Recently, police tracked Williams Amaya to his place of employment, right after he allegedly shot his aunt and uncle over a disagreement about a dog.
Their cell phones gave them away.
How it works
The digital world in which we live generates massive amounts of data. To find you or your cell phone, that data is broken down into pieces you can talk to, said Don Reich, co-founder of the Public Safety Network.
Your smart phone usually has a GPS on it, but cell towers don’t always point you in the right direction. So they’re now able to figure out where you are by triangulating them off a few different cell towers, Reich explained.
Service providers can “ping” a signal to a cell phone and have it respond. The term stems from SONAR when a technician would send out a sound wave, or ping, and wait for its return to locate another object.
Tracking the data
If you think 911 phone data are being analyzed, you’re right, says Brian Fontes with NENA, the National Emergency Number Association.
“The goal is to take care of their needs at the time of their greatest need,” Fontes said.
Across the U.S., emergency responders take 240 million 911 calls each year, said Jamie Barnett with Findme911.com.
Of those, 70-80 percent calls are from wireless devices, around 168 million.
“That’s 168 million people on the most important phone call of their lives,” Barnett said. “People claim that the government is trying to track them. No. You called 911 and you want to be found.”
Your 911 call will go to one of the country’s 135,000 911 operators.
In Eagle County, emergency calls goes through a central dispatch center based in Vail.
In Denver, 911 director Carl Simpson knows exactly how many calls they get and when they roll in. Denver has 66 people each watching five or six screens take those calls, and almost every single one of them is answered on the first ring, Simpson said.
“It’s a busy shop,” Simpson said.
They know how many calls they’ve gotten on any given day for years at a time.
“I’ll bet that on the 45th Monday of this year we get just about that many calls,” Simpson said.
The data changed the way they do business, Simpson said.
They used to have an equal number of staffers on each shift, Simpson said.
“Guess what? Calls don’t come in like that,” Simpson said.
What they do with the data
Roger Hixon analyzes the data for NENA, looking for patterns. He found several.
“In other words, how often do people call because their lives are in danger, and how many call because they need a cab and are too drunk to dial more than three numbers?” Hixon asked.
It’s also used to help law enforcement anticipate crime patterns based on what they’ve see in the past.
It’s called Predictive Analytics, and it’s being used on you already. Amazon uses it to predict whether you might like something and want to buy it, Hixon said.
“Information is not knowledge. You have to take something away from the information that you can apply,” said Paul McLaren, director of support engineering at Intrado.
It’s also called Predictive Policing, and sometimes it’s incredibly complex. But sometimes it’s as simple as flipping a light switch, said Alex Kreilein with the University of Colorado’s Information and Technology labs.
“A shooting occurred at this one place about every other day,” Kreilein said. “Instead of staking out five or ten police cars, they took another look. It turns out that even though it was getting dark earlier in the fall, the streetlights still didn’t go on until 7 p.m. When it’s dark, more crime occurs, so they turned the lights on earlier. That actually solved most of it.”
80/20 crime stats
In Rochester, New York, law enforcement has been compiling data on criminals in their town, Hixon said. It’s based on the 80/20 rule: 20 percent of the bad guys will commit 80 percent of the crime. Once they compiled that list they put it in the hands of cops on the street, along with data about the patterns of behavior cops can expect to encounter when they arrive on the scene.
Take property crimes, for instance. Once the police have their frequent flyer list of criminals, they can assign data to how the potential perpetrators tend to behave — break in the back door, steal televisions — those sorts of behaviors.
After a series of property crimes, the Rochester police worked down their frequent flyer list and started knocking on doors. No. 5 on the list was their winner and was also nailed for 24 other property crimes, Hixon said.
In Oceanside, Calif., some scoundrel stole an Olympic gold medal. There was just the one gold medalist in Oceanside, so it was a big deal. The police used a bunch of data mined from past crimes to catch the guy and get the gold medalist his medal back.
There’s also a category that assigns probabilities to how likely those crimes are to be gang related.
‘More bureaucracy than technology’
Dianna Anderson is Colorado’s state chief data officer with the governor’s Office of Information Technology.
“The biggest problem is getting the agencies to agree to share data. It’s more bureaucracy than technology. They’re very protective of their information, as they should be,” Anderson said. “Analytics is so much more powerful if you can get agencies to share information.”
Support Local Journalism
Support Local Journalism
Readers around Glenwood Springs and Garfield County make the Post Independent’s work possible. Your financial contribution supports our efforts to deliver quality, locally relevant journalism.
Now more than ever, your support is critical to help us keep our community informed about the evolving coronavirus pandemic and the impact it is having locally. Every contribution, however large or small, will make a difference.
Each donation will be used exclusively for the development and creation of increased news coverage.
Start a dialogue, stay on topic and be civil.
If you don't follow the rules, your comment may be deleted.
User Legend: Moderator Trusted User