Big Data // Big Data Analytics
Commentary
1/21/2010
04:55 PM
Rob Preston
Rob Preston
Commentary
Connect Directly
LinkedIn
Twitter
RSS
E-Mail
50%
50%

Down To Business: In Defense Of Profiling

While profiling is often associated with police work, everyone from insurance companies to bank lenders to e-commerce sites to fast food restaurants engages in the practice.

InformationWeek Green - January 25, 2010 InformationWeek Green
Download the entire January 25, 2010 issue of InformationWeek, distributed in an all-digital format as part of our Green Initiative
(Registration required.)
We will plant a tree
for each of the first 5,000 downloads.

Few subjects raise the hackles of civil libertarians and ordinary citizens alike more than profiling, the process by which institutions target people based on traits such as race, national origin, ethnicity, psychological makeup, behavior, and past actions.

My cousin Jake McNicholas, a retired New York City police detective, was roundly criticized, for instance, after he wrote a recent newspaper column describing how he often had "used a person's race to initiate investigations and make subsequent arrests." The tools of his trade? Mere observation: following white guys driving cars with New Jersey plates slowly through predominantly black neighborhoods, in many cases looking for drugs. Even when profiling works and is done without malice, the fear of abuse is a powerful undercurrent.

While profiling is most often associated with cops, everyone from insurance companies to bank lenders to e-commerce sites to fast-food restaurants engages in the practice. The tools of their trade? Data mining and analytics software, tracking and monitoring technology, and other kinds of IT. Done right, these tools help companies identify the best potential customers or help them give customers precisely what they want when they want it. Done wrong, they invade privacy and can introduce serious mistakes. And there's a sea of gray between the right and wrong.

The police and intelligence communities are also big IT users, with mixed success and public backing. One of IBM's chief scientists, Jeff Jonas, has warned against using data mining to root out terrorists in the broad population. Jonas has argued that data mining is useful in identifying shopping habits and financial fraud, where results don't have to be pinpoint accurate, but not in identifying a few potential terrorists among hundreds of millions of people. Such analyses register false positives more than 90% of the time, Jonas argued in a report written a few years ago with the Cato Institute's Jim Harper, rendering analytics for anti-terrorism "useless and potentially harmful." Data mining has had success in law enforcement on a far more limited scale, like with deploying more police to areas where data analysis has determined an increased probability of a crime occurring.

I got to thinking more about IT-based profiling after reading a fascinating column by my colleague Alex Wolfe on a dozen recent IBM patent applications for airport security technology that purports to analyze characteristics such as age, type of clothing, facial expressions, even smell to identify potential terrorists. The applications outline networks of video, motion, chemical, and biometric sensors at airports that would feed into a local computer grid, yielding results in near real time. As described by Alex, the key technology IBM is seeking to patent is a software "inference engine" that uses heuristics and rules developed by the three co-inventors to crunch the data and sort out high-risk people.

One of the inventors, Robert Angell, contacted Alex, after Alex had written an initial blog post on the subject, to emphasize that if the technology is applied correctly, "we could do passive profiling [and] passive detection and do it without a whole lot of fanfare." Of course, whenever people are singled out based on features such as age and facial nuances, there's bound to be a hue and cry. (Ask my cousin Jake.) As Alex points out, while the IBM inventors aren't playing up the profiling aspect of their work, "it's clearly a significant piece of what they're doing."

But don't rule out profiling (and tracking and monitoring) just because someone somewhere conjures up George Orwell. In a country where Norwegian grandmothers are searched at airports about as often as individuals who more closely fit the established terrorist profile--in the name of fairness, not effectiveness--we might consider applying technology that could do a better job.

Rob Preston,
VP and Editor in Chief, InformationWeek
rpreston@techweb.com

To find out more about Rob Preston, please visit his page.

Comment  | 
Print  | 
More Insights
6 Tools to Protect Big Data
6 Tools to Protect Big Data
Most IT teams have their conventional databases covered in terms of security and business continuity. But as we enter the era of big data, Hadoop, and NoSQL, protection schemes need to evolve. In fact, big data could drive the next big security strategy shift.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest Septermber 14, 2014
It doesn't matter whether your e-commerce D-Day is Black Friday, tax day, or some random Thursday when a post goes viral. Your websites need to be ready.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.