Fuck you, spam

2008-07-09 Wed – 13:20:07

And this, folks, is why spam sucks so fucking much:


Junk Filter Statistics

The junk mail filter has been trained by 46868 messages, whereof 28127 (60%) have been rated as solicited and 18741 (40%) as junk. This resulted in a total of 699360 tokens read, 349780 (50%) rated as good and 349580 (50%) as evil; the number of different tokens is 622268.
The following table will show the 12 most common tokens, hiding 622256 tokens below the threshold of 24060 appearances.

(Processing this training.dat of 24060562 bytes took 780.35 seconds.)

Token                            Good   Evil  Junk Probability
1 mime-version:1.0              13818  17248  65.20 %
2 for                           16579   9394  45.96 %
3 the                           21511  11836  45.23 %
4 envelope-to:pgl@yoyo.org      22742  16294  51.81 %
5 content-type/type:text/plain  22016   6247  29.87 %
6 you                           19875   9279  41.20 %
7 with                          17209   8710  43.17 %
8 this                          17078   9004  44.17 %
9 x-mozilla-status2:00000000    10897  13810  65.54 %
10 that                         17419   7984  40.76 %
11 x-mozilla-status:0001        10846  13810  65.65 %
12 and                          20086  10530  44.03 %

 

That’s right, there’s a greater than 40% chance any mail I receive is going to be spam if it contains any of these words: “with”, “that”, “and”, “for”, “the”, or “you”. 40% for any of them. Sigh.

Sorry, comments for this entry are closed at this time.