[Python-Dev] Re: The first trustworthy <wink> GBayes results
28 Aug 2002 14:10:54 -0000
Bayesian filters are pretty robust in the face of corpus
contamination, if you have a threshold for the number of
occurrences of a word that you'll consider. If you don't
do that, then yes, a single legit email in your spam
corpus could cause your filters to reject every similar
A single email could easily contain five to eight words
that never occur in any other email. (Username, domain
name, server name, street address, etc.) If this got
into your spam corpus by mistake, then every succeeding
email from the same person would be classified as spam.
What this means is that you may want to use slightly
different thresholds for occurrences depending on how
much you trust the (human) classifier. For an app to be
used by end users, you might want to have a high threshold,
like 20 occurrences.
I find from my own experience that I often misclassify
mail. I seem to be more likely to put spam in a legit
mail folder than the reverse. But, as you guys found,
the first result of testing your filters tends to be to
clean up such mistakes.
--Greg Ward wrote:
> On 27 August 2002, Tim Peters said:
> > Setting this up has been a bitch. All early attempts floundered because it
> > turned out there was *some* systematic difference between the ham and spam
> > archives that made the job trivial.
> > The ham archive: I selected 20,000 messages, and broke them into 5 sets of
> > 4,000 each, at random, from a python-list archive Barry put together,
> > containing msgs only after SpamAssassin was put into play on python.org.
> > It's hoped that's pretty clean, but nobody checked all ~= 160,000+ msgs. As
> > will be seen below, it's not clean enough.
> One of the other perennial-seeming topics on spamassassin-devel (a list
> that I follow only sporodically) is that careful manual cleaning of your
> corpus is *essential*. The concern of the main SA developers is that
> spam in your non-spam folder (and vice-versa) will prejudice the genetic
> algorithm that evolves SA's scores in the wrong direction. Gut instinct
> tells me the Bayesian approach ought to be more robust against this sort
> of thing, but even it must have a breaking point at which misclassified
> messages throw off the probabilities.
> But that's entirely consistent with your statement:
> > Another lesson reinforces
> > one from my previous life in speech recognition: rigorous data collection,
> > cleaning, tagging and maintenance is crucial when working with statisical
> > approaches, and is damned expensive to do.
> On corpus collection...
> > The spam archive: This is essentially all of Bruce Guenter's 2002 spam
> > collection, at <http://www.em.ca/~bruceg/spam/>. It was broken at random
> > into 5 sets of 2,750 spams each.
> One possibility occurs to me: we could build our own corpus by
> collecting spam on python.org for a few weeks. Here's a rough breakdown
> of mail rejected by mail.python.org over the last 10 days,
> eyeball-estimated messages per day:
> bad RCPT 150 - 300 
> bad sender 50 - 190 
> relay denied 20 - 180 
> known spammer addr/domain 15 - 60
> 8-bit chars in subject 130 - 200
> 8-bit chars in header addrs 10 - 60
> banned charset in subject 5 - 50 
> "ADV" in subject 0 - 5
> no Message-Id header 100 - 400 
> invalid header address syntax 5 - 50 
> no valid senders in header 10 - 15 
> rejected by SpamAssassin 20 - 50 
> quarantined by SpamAssassin 5 - 50 
>  this includes mail accidentally sent to eg. email@example.com,
> but based on scanning the reject logs, I'd say the vast majority
> is spam. However, such messages are rejected after RCPT TO,
> so we never see the message itself. Most of the bad recipient
> addrs are either ancient (firstname.lastname@example.org,
> email@example.com) or fictitious (firstname.lastname@example.org,
>  sender verification failed, eg. someone tried to claim an
> envelope sender like email@example.com. Usually spam, but innocent
> bystanders can be hit by DNS servers suddenly exploding (hello,
> comcast.net). This only includes hard failures (DNS "no such
> domain"), not soft failures (DNS timeout).
>  I'd be leery of accepting mail that's trying to hijack
> mail.python.org as an open relay, even though that would
> be a goldmine of spam. (OTOH, we could reject after the
> DATA command, and save the message anyways.)
>  mail.python.org rejects any message with a properly MIME-encoded
> subject using any of the following charsets:
> big5, euc-kr, gb2312, ks_c_5601-1987
>  includes viruses as well as spam (and no doubt some innocent
> false positives, although I have added exemptions for the MUA/MTA
> combinations that most commonly result in legit mail reaching
> mail.python.org without a Message-Id header, eg. KMail/qmail)
>  eg. "To: all my friends" or "From: <>"
>  no valid sender address in any header line -- eg. someone gives a
> valid MAIL FROM address, but then puts "From: firstname.lastname@example.org"
> in the headers. Easily defeated with a "Sender" or "Reply-to"
>  any message scoring >= 10.0 is rejected at SMTP time; any
> message scoring >= 5.0 but < 10 is saved in /var/mail/spam
> for later review
> Executive summary:
> * it's a good thing we do all those easy checks before involving
> SA, or the load on the server would be a lot higher
> * give me 10 days of spam-harvesting, and I can equal Bruce
> Guenter's spam archive for 2002. (Of course, it'll take a couple
> of days to set the mail server up for the harvesting, and a couple
> more days to clean through the ~2000 caught messages, but you get
> the idea.)
> > + Mailman added distinctive headers to every message in the ham
> > archive, which appear nowhere in the spam archive. A Bayesian
> > classifier picks up on that immediately.
> > + Mailman also adds "[name-of-list]" to every Subject line.
> Perhaps that spam-harvesting run should also set aside a random
> selection of apparently-non-spam messages received at the same time.
> Then you'd have a corpus of mail sent to the same server, more-or-less
> to the same addresses, over the same period of time.
> Oh, any custom corpus should also include the ~300 false positives and
> ~600 false negatives gathered since SA started running on
> mail.python.org in April.