Re: [Evolution] Spam Filtering

Hans wrote:
A hack-ish workaround to train Bayes using Evo is the following: Pick at
least 200 non-Junk messages, and keep them in mind. Now mark them as
Junk using the Evo UI. Go to your Junk folder, and mark all those
non-Junk mail we just abused in a sacrificial manner and correctly mark
them as non-Junk. SA will realize it learned these messages previously,
and learn them as Ham (non-Junk) only, AFAIK. Now we got the 200 Hams
learned. Collecting at least 200 Spams for learning shouldn't be hard I
guess. ;-)

Note: This really is a *hack* only, and you should not try this unless
you feel a little bit adventurous. :)

Do I have to mark the mail as Junk first and then un-Junk them - or can
I just mark them in the Inbox as Not-Junk? In other words, what does the
initial marking as Junk do?
When you mark an email as Spam, it enables Spamassassin to analyse the email's various characteristics and add it to its paramjeters that identify spam - the more times you train it with different emails the more confident Spamassassin can be in identifying Spam itself - pretty much how we learn to decide whether we like or don't like people we meet, based on past experience. When you mark an email as not spam, when Spamassassin has decided it is spam, you are forcing it to rethink some rules where it had perhaps guessed wrong. For instance, somebody sends you a learned treatise on the medical effects of Viagra, Spamassassin may werll treat it as spam, but if you tell it its's not spam, it will go away and think "not all references to Viagra mean spam, let's think more deeply about this, maybe I need to modify my rules".
Evolution-list mailing list
Evolution-list gnome org

------------------------------------------------------------------------------------------------ - find out how to make your computer work for the community

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]