When a new release comes out I like to be anal-retentive and go through the
GA second-guessing its scores. This is my report for 2.30.

Overall, the GA did a NICE job this time. I have very little to complain
about and haven't found a single score I'll be bothering to override. Here
are a few scores I think came out really well:

- Advert codes are, of course, a good indicator
score ADVERT_CODE                    4.725

- Spam phrases are actually turning out useful this time
score FREQ_SPAM_PHRASE               2.417
score SPAM_PHRASES_020               2.139
score SPAM_PHRASES_040               2.424

- New uppercase frequency checks work well
score UPPERCASE_25_50                1.937
score UPPERCASE_50_75                2.972
score UPPERCASE_75_100               2.990

- Future dates are a surprisingly good spam indicator
score DATE_IN_FUTURE_03_06           3.416
score DATE_IN_FUTURE_06_12           2.385
score DATE_IN_FUTURE_12_24           3.308
score DATE_IN_FUTURE_24_48           3.657
score DATE_IN_FUTURE_48_96           2.887
score DATE_IN_FUTURE_96_XX           3.463

- RATWARE must be fixed, it was negative last time
score RATWARE                        4.563

- This works well for me but users in some countries may want to change it
score SUBJ_FULL_OF_8BITS             4.298

And a few slightly questionable scores:

- This was 0.87 before. Less and less useful?
score FROM_AND_TO_SAME               -2.071

- Not as weird as all that, apparently
score MSGID_CHARS_WEIRD              -2.178

- Disappointing, perhaps porn_word_test() needs tweaking
score PORN_3                         0.522

- Lots of missing dates in non-spam?
score DATE_MISSING                   -2.140

Just for the record, here's the usual list of rules that ended up with
slightly negative scores. They aren't really good rules for catching
nonspam, so I think the rules are likely either defective or obsolete.

score ASCII_FORM_ENTRY               -1.660
score ASKS_BILLING_ADDRESS           -0.152
score DEAR_SOMEBODY                  -0.694
score EXCUSE_16                      -0.721
score FORGED_HOTMAIL_RCVD            -0.356
score FROM_NAME_NO_SPACES            -0.114
score GREEN_EXCUSE_1                 -2.019
score INTL_EXEC_GUILD                -0.039
score LINES_OF_YELLING               -0.036
score MONEY_BACK                     -0.239
score MONEY_MAKING                   -0.687
score NO_REAL_NAME                   -1.068
score SUBJ_ALL_CAPS                  -0.054
score SUBJ_ENDS_IN_Q_MARK            -0.135
score SUBJ_REMOVE                    -0.823
score SUSPICIOUS_RECIPS              -0.213
score WEB_BUGS                       -0.430
score X_AUTH_WARNING                 -0.703
score X_ESMTP                        -1.662
score X_MSMAIL_PRIORITY_HIGH         -0.886
score X_NOT_PRESENT                  -1.920
score MAILTO_TO_REMOVE               -1.669

All in all, I believe the GA is really smarter than I am this time. :)

--
michael moncur   mgm at starlingtech.com   http://www.starlingtech.com/
"My one regret in life is that I am not someone else."    -- Woody Allen


_______________________________________________________________

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas - 
http://devcon.sprintpcs.com/adp/index.cfm?source=osdntextlink

_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk

Reply via email to