Data mining: the key to unlock your spending patterns

I have a good friend who prides himself on flying below the internet radar. He does everything he can think of to keep his name out of the press and off any official documents which may end up in the public domain. He doesn't use his real name when registering for web sites and services. Even his personal email address bears no resemblance to his own name. Indeed a search for his name on any of the major search engines turns up nothing appearing to have any relevance to my friend at all.

He does it because he says he values his privacy. He doesn't want people to find out things about him by entering his name in search engines. He believes the less his name turns up the better his chances for not falling victim to some form of fraud or another.

If only he realized how naive his pursuit is.

He may be keeping his own name out of the public eye, and that may be a successful tactic for keeping the merely curious away from his affairs; but there is little he can do to prevent those with a profit motive from finding him.

They may not find him in the traditional sense, but they'll find him in the aggregate. Whether he likes it or not he's part of any number of groups of people with similar characteristics or behaviors. Using data mining technologies, corporations can devise strategies to put just the right offer in front of people in his group and, with a reasonable likelihood, watch him take the bait.

Data mining techniques sort through enormous amounts of data on nearly every circumstance surrounding certain events. The idea is to determine what combinations of conditions exist where a particular outcome is more likely. You may have heard stories about grocery stores putting diaper displays near beer promotions. The practice was devised from a crude and early form of data mining which determined that many diaper purchases were made by men.

Fast forward to today and you'll find computers looking at sales data and correlating it to not only the time of day but also the weather and your city's major league baseball team's home and away schedule. They're also looking into correlations between buying trends and the things people buy along with the things they buy - not just this purchase, but purchases yesterday and last week and last month.

If my friend really wanted to fly under the radar he'd do well to forgo use of his credit and debit cards altogether. It wouldn't completely accomplish his goal however. He'd still be participating in some stream of data gathering - those who purchased X, Y and Z items with cash on a sunny Wednesday afternoon in June when the Cubs were on the road and gas prices had just dipped a few cents.

Data Mining is a fascinating subject. If you're like me and enjoy learning about the techniques and processes behind new technologies you might want to consider one of these books on the topic of Data Mining at Amazon.com.