Using Big Data to Predict Crime Patterns
What You Need to Know about the ISA 101 HMI Standard: Empowering the Operator
October 30, 2015Looking Back on 2015: A Year in Review
December 8, 2015If data experts are using big data to predict weather patterns, why would forecasting crime be any different?
According to a recent BBC iWonder post, police in both Los Angeles and Manchester ran similar trials using a computer algorithm to predict where crime would take place. Their goal was to find patterns in criminal behavior by analyzing large amounts of crime data in hopes to eventually prevent crime– a term called “predictive policing.”
Here’s how it works, according to the BBC report.
Using a hypothetical case study based on actual patterns found in the data, researchers at University College London pinpoint a location where a crime takes place– for example, a house is burgled in a suburb.
The risk of more burglaries now extends to houses in the immediate area, with the highest threat level within about 220 yards. In fact, research showed that the risk is greater on the same side of the road where the original burglary occurred. The risk of further crime continues within two weeks, but dissipates as time goes on.
Let’s say another crime happens in the same area. The risk now spreads out from both the original location and the new one– meaning police can now focus specifically where they intersect.
Now, if predictive policing sounds like a familiar movie plot–the 2002 film, “Minority Report,” anyone?– don’t panic.
According to Slate Magazine, Jeffrey Brantingham, anthropology professor at UCLA and partner to the LAPD research team, assures us it’s nothing like Tom Cruise’s character arresting people for “pre-crime.”
“This is not about predicting the behavior of a specific individual,” said Brantingham in the article. “It’s about predicting the risk of certain types of crimes in time and space.”
Christopher Beam, author of the Slate article, said that the idea behind predictive policing is that while some crime is random, a lot is not.
And while plenty of people think that a burglarized house is like like lightning– it never strikes twice– the previous example illustrates the exact opposite. In many cases, houses in close proximity are at a higher risk.
[ut_alert color=”grey”] “That doesn’t mean police can prevent every kind of crime—only the more predictable kinds, like burglary or auto theft,” said Beam. “And even then, some robberies could be truly random. But predictive policing could reduce crime on the margins, according to the UCLA researchers.” [/ut_alert]
So, does it work? According to the BBC, in instances where specific patterns are exhibited, it’s helping to cut down on crime. In fact, Manchester police noticed a 26.6% decline in burglaries in 2011.
Of course, more trials are needed for an even more accurate analysis. According to the article, Rachel Tuffin, director of research at the College of Policing said there are plans for further research, and we could expect more published findings this year.
And here are some other notable stories from this week:
Big Data is Now a Top Management Issue
According to a Forbes article from this past week, a new report from Economist Intelligence Unit has shown that big data is “moving from its infancy to ‘data adolescence,’ in which companies are increasingly meeting the challenges of a data-driven world.”
The report said that in the past 5 years, more and more companies have begun treating their data as a “strategic corporate asset.” And not just any data, the data that can help solve problems and address certain areas.
Because of this, the report said that data strategy is becoming much more of a priority for those in a leadership position.
“Data strategy has been elevated to the C-level, usually centralized with a CIO/CTO or a newly-appointed Chief Data Officer (CDO). Outside that position, executives across the board are more likely to be in charge of their departments’ particular data initiatives and instrumental in putting those resources to use.”
The Internet of Things Market
According to a recent press release, the industrial space has a hold on the top market share in the Internet of Things (IoT) market.
Here’s some fast facts from the report:
- The healthcare and consumer electronics sector is the fastest growing in the IoT market
- In 2014, the automotive industry is the second largest application segment in the global IoT market
- North America represents the largest market share, and is the largest revenue share in the global market
- The rapid growth of industrial, automotive and healthcare industries is the major factors driving the growth in North America
Read the full report here.
Raspberry Pi’s $5 Computer Sold Out in Flash
Raspberry Pi, the credit card-sized, single-board computer series, has been pushing down the cost of computers for years in the pursuit of teaching basic computer science in an affordable way.
But over Thanksgiving weekend the foundation released their newest model, Raspberry Pi Zero, at an all-time low price of $5– roughly the price of a large latte.
It shouldn’t come as too big of a surprise then that the model sold out in the same day, according to Popular Mechanics. That’s about 20,000 of the initial run.
”You’d think we’d be used to it by now, but we’re always amazed by the level of interest in new Raspberry Pi products,” Eben Upton, founder of the Raspberry Pi Foundation, told Wired.
Media We Link To:
“Can big data help us predict where crime will strike?” – BBC
“Time Cops: Can police really predict crime before it happens?” – Slate Magazine
“Big Data: Now A Top Management Issue” – Forbes
“Raspberry Pi’s Dirt-Cheap $5 Computer Sold Out in a Flash” – Popular Mechanics
