After playing with a Perl implementation for a while, I suspect that Graham simplified his implementation for purposes of writing the article. In particalar, the examples he shows suggest that he unconditionally converts tokens to lower case. I did this when transliterating his code, but found it failed when faced with short spams that read something like 25 MILLION EMAILS! http://... Both "million" and "emails" weight in below 0.5. But when I use a scheme that lowercases only mixed-case tokens (giving "million" and "MILLION" separate weights), I get fewer false negatives on spam. The problem is that the weighting tables gets a lot larger. Mine, drawn from a sample of about 2,000 spam and 1,500 non-spam emails, is 1.5Mb. That's rather large to be scanning when doing one-off email filtering (e.g., via procmail).

In reply to Re: Bayesian Filtering for Spam by dws
in thread Bayesian Filtering for Spam by oakbox

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.