Why do most spam messages contain a bunch of random words?
As Michelle Delio’s excellent article from Wired News explains, those chunks of gibberish are intended to trick spam filters. More specifically, they’re meant to disrupt spam filters that use Bayesian analysis techniques. Named after 18th-century minister and statistician Thomas Bayes, Bayesian techniques are used by programs to search for words and phrases that most likely appear in a particular user’s legitimate email. Spammers throw in random words (along with the predictable sales pitch) in order to trick the filter into thinking that the email is personal correspondence. But to really fool a Bayesian filter, the gibberish (commonly called a
