December 12, 2005

Decrapifying User-Generated Content

Earlier, Greg Linden brought up the issue of filtering junk "UGC". Now, Ian McAllister of MSN gives a more detailed overview of filtering techniques.

Speaking of bad words, comparison engines might do well to flag words like N*geria (1) (2) for manual review...

Decrapifying User-Generated Content
Posted by Sean O'Rourke on December 12, 2005 at 12:49 PM
Archived at Product Reviews