bayes poisoning

[maillist <ma...@emailacs.com>]

I see a few emails every-now-and-then about "bayes poisoning", and am 
wondering what is means.  From what I understand, it is some message 
that gets learned (only through autolearn?) that has certain 
characteristics that throw the bayes system off.

The reason I ask, is I got the following email...

/Shields cameron diaz, carmen, electra, catherine zeta jones.
Spears brooke burke burns, shields cameron. Rai, alanis morissette, alesha oreskovich, ali landry. Berry heidi, klum hilary duff jennifer.
Lindsay lohan liv tyler, madonna mandy mariah.

Company: CHINA BIOLIFE ENTERP (Other OTC:CBFE.PK)
Symbol: CBFE
Price: $1.55
Target: $4
Market: Bullish

Hilary duff jennifer aniston lopez love hewitt jessica alba?
Mandy mariah, carey, mary meg ryan milla, jovovich. Celebossru, playboy models new, home aaliyah.
Halliwell gillian anderson gwen stefani halle berry heidi.
Semenovich, aria giovanni ashlee simpson. Farmer natalia oreiro natalie, imbruglia portman. Ashlee simpson ashley olsen audrey hepburn, avril lavigne, britney? Kidman, twins pamela, paris hilton penelope cruz peta wilson! Halliwell gillian anderson gwen stefani halle berry heidi. Silvstedt winona ryder, zdenka podkapova photo./


This doesn't seem to be any sort of "spam" per se, but I didn't know if 
it was an attempt to "poison" anything.

....cause if I miss an email from Jennifer Aniston, or any playboy 
models, I will not be able to recover.  :)


-=Aubrey=-

Re: bayes poisoning

[Chris Purves <ch...@northfolk.ca>]

maillist wrote:
> I see a few emails every-now-and-then about "bayes poisoning", and am 
> wondering what is means.  From what I understand, it is some message 
> that gets learned (only through autolearn?) that has certain 
> characteristics that throw the bayes system off.
> 

 From what I've seen there are generally two ways it is referred to:

1. random text or phrases thrown into spam to make it look like spam and 
ham look more alike:  This is an imagined problem.

2. spam incorrectly leanred as ham or ham incorrectly learned as spam: 
Enough of these (either from manual or auto learning) and your Bayes 
database will be useless.

-- 
Chris