Bayes still effective?

Raymond Dijkxhoorn raymond at PROLOCATION.NET
Wed Jul 28 22:46:15 IST 2004


Hi!

> MessageI'm not using SpamAssassin's Bayes yet, but considering implementing
> it.   As you all know, the problem is that most spam deliberately tries to
> mess up the Bayes engine by including lists of words buried in the message.
>
> This strikes me as a pretty effective way of circumventing bayes.  Is it
> worth bothering with?

Its VERY effective, still... todays stats:

SpamAssassin tag hits: (top 100)
#1      202469  BAYES_99
#2      150136  RCVD_IN_SBL+XBL
#3      145564  HTML_MESSAGE
#4      137984  WS_URI_RBL
#5      137831  RCVD_IN_BL_SPAMCOP_NET
#6      137514  OUTBLAZE_URI_RBL
#7      118086  RCVD_IN_SORBS
#8      91947   RCVD_IN_DSBL
#9      90897   SPAMCOP_URI_RBL
#10     85217   ABUSEBUTLER_URI_RBL
#11     84117   MIME_HTML_ONLY
#12     62749   RCVD_IN_DYNABLOCK
#13     62418   RCVD_IN_AHBL
#14     56386   CLICK_BELOW
#15     54672   RCVD_IN_NJABL
#16     45978   DNS_FROM_RFCI_ABUSE
#17     43563   MIME_HTML_ONLY_MULTI
#18     39290   HTML_FONT_BIG
#19     39248   LOCAL_XMESSAGEINFO
#20     38787   HTML_LINK_CLICK_HERE
#21     38219   MIME_HTML_NO_CHARSET
#22     37462   MSGID_FROM_MTA_HEADER
#23     32530   BIZ_TLD
#24     29231   RCVD_IN_RFCI
#25     27232   DRUGS_ERECTILE
#26     27194   RCVD_IN_NJABL_PROXY
#27     25757   HTML_60_70
#28     21999   RCVD_IN_SORBS_SOCKS
#29     21856   HTML_FONTCOLOR_BLUE
#30     21835   RCVD_IN_SORBS_HTTP
#31     21183   HTML_70_80
#32     20614   HTML_FONTCOLOR_RED
#33     20423   RCVD_IN_OPM
#34     18717   HTML_50_60
#35     17543   FROM_ENDS_IN_NUMS
#36     15618   NO_REAL_NAME
#37     15560   HTML_FONTCOLOR_UNSAFE
#38     15059   HTML_MIME_NO_HTML_TAG
#39     14781   J_BACKHAIR_22
#40     14599   RCVD_IN_OPM_HTTP
#41     13400   J_BACKHAIR_23
#42     13357   HTML_FONTCOLOR_UNKNOWN
#43     13328   HTML_FONT_INVISIBLE
#44     13308   MISSING_MIMEOLE
#45     13245   RCVD_IN_NJABL_SPAM
#46     13221   OFFERS_ETC
#47     13198   RCVD_IN_NJABL_DIALUP
#48     12977   DRUGS_ERECTILE_OBFU
#49     12974   J_BACKHAIR_12
#50     12538   DRUGS_PAIN
#51     12159   J_BACKHAIR_24
#52     11849   HTTP_WITH_EMAIL_IN_URL
#53     11810   HTML_IMAGE_ONLY_04
#54     11632   FORGED_YAHOO_RCVD
#55     11521   UPPERCASE_25_50
#56     11446   DNS_FROM_RFCI_DSN
#57     11372   RCVD_IN_OPM_SOCKS
#58     11370   HTML_WEB_BUGS
#59     10590   J_BACKHAIR_32
#60     10161   J_BACKHAIR_13
#61     10031   J_BACKHAIR_11
#62     10018   J_BACKHAIR_43
#63     9861    HTML_IMAGE_ONLY_02
#64     9366    DRUGS_ANXIETY
#65     9339    J_BACKHAIR_33
#66     9270    J_BACKHAIR_14
#67     9268    PRIORITY_NO_NAME
#68     9140    REMOVE_PAGE
#69     9015    MIME_BOUND_NEXTPART
#70     8942    DRUGS_MANYKINDS
#71     8568    DRUGS_MUSCLE
#72     8547    J_BACKHAIR_21
#73     8533    DRUGS_PAIN_EREC
#74     8352    FORGED_HOTMAIL_RCVD2
#75     8170    MIME_HEADER_CTYPE_ONLY
#76     8111    DRUGS_DIET
#77     7836    RCVD_IN_OPM_HTTP_POST
#78     7801    NORMAL_HTTP_TO_IP
#79     7773    J_BACKHAIR_31
#80     7771    DRUGS_ANXIETY_EREC
#81     7760    ALL_NATURAL
#82     7723    HTML_TITLE_EMPTY
#83     7490    J_BACKHAIR_41
#84     7406    DRUGS_DIET_EREC
#85     7217    EXCUSE_14
#86     7152    HTML_IMAGE_ONLY_10
#87     7132    MAILTO_SUBJ_REMOVE
#88     6978    HTML_IMAGE_ONLY_06
#89     6911    J_BACKHAIR_44
#90     6835    MISSING_OUTLOOK_NAME
#91     6722    MIME_BASE64_TEXT
#92     6685    RCVD_NUMERIC_HELO
#93     6623    J_BACKHAIR_42
#94     6455    FORGED_OUTLOOK_TAGS
#95     6436    SORTED_RECIPS
#96     6293    ONLINE_PHARMACY
#97     6293    J_BACKHAIR_34
#98     6275    SUSPICIOUS_RECIPS
#99     6107    FORGED_MUA_OUTLOOK
#100    5983    HTML_IMAGE_RATIO_04

Bye,
Raymond.

-------------------------- MailScanner list ----------------------
To leave, send    leave mailscanner    to jiscmail at jiscmail.ac.uk
Before posting, please see the Most Asked Questions at
http://www.mailscanner.biz/maq/     and the archives at
http://www.jiscmail.ac.uk/lists/mailscanner.html



More information about the MailScanner mailing list