Data Mining the AOL Privacy Blunder
Posted August 16, 2006 10:24 AM
Folllowing up on my AOL Eternal Hell posting, it appears that Lee Gomes has downloaded the 2.27 gigabytes of search data to discover how people really use the web.
Most commonly used word: "free"; #2 - "new"; followed by "lyrics," "county," "school," "city," "home," "state," "pictures," "music," "sale," "beach," "high," "map," "center" and "sex."
47% of all searches do not result in a "click"
22% (42% of 53%) clicked the first result
28% are refinements of prior searches
15% were domain names meant for the address bar [Valleywag: "Great, now we have quantified evidence that AOL means Low IQ."] My conclusion is that this is perfectly acceptable when you are seeking the Google cache of the url, as I often do...
[WSJ Free Features: What Are Web Surfers Seeking? Well, It's Just What You'd Think]
Note: Pete Wentz of Fall Out Boy is most searched person on the net.
See also: John Walkenbach at J-Walk Blog (Walkenbach wrote Excel 2007 Bible) which is why he put this AOL data leak into a handly Excel Spreadsheet.
Related: Data Mining Enron Emails
About
This entry, Data Mining the AOL Privacy Blunder, is part of the Tech Law Advisor weblog and was posted by in the research Category. You can subscribe to this weblog via RSS with this feed. If you plan on using this content for any purpose, you should review this disclaimer and copyright notice.