Re: training bayes and newsletters

Bill Cole <[email protected]>
Newsgroups gmane.mail.spam.spamassassin.general
Message-ID <[email protected]>
On 2024-10-15 at 22:05:07 UTC-0400 (Tue, 15 Oct 2024 22:05:07 -0400)
Alex <[email protected]>
is rumored to have said:

> I can imagine the newsletter template is somewhat common, but does 
> bayes
> have any ability to distinguish a junk newsletter from a legitimate
> newsletter?

Not if it has never seen either of them.

Bayesian classification is not magic, but it does do better than random 
guessing  by someone who is not the target of the message, IF it has 
been trained on similar mail that has been properly classified.

> I realize there's somewhat of an imbalance between hams and
> spams, but shouldn't there be enough?

Absolutely. But the Bayes classifier can't classify mail of types that 
have been meticulously excluded from its training corpus.

> Would I benefit from training known trustworthy newsletters such as 
> ham?

Yes. And train the spam ones as spam.

-- 
Bill Cole
[email protected] or [email protected]
(AKA @[email protected] and many *@billmail.scconsult.com 
addresses)
Not Currently Available For Hire
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.