Mailing List Archive

svn commit: r1914758 [2/2] - in /spamassassin/trunk/rulesrc/scores: 72_scores.cf scores-set0 scores-set1 scores-set2 scores-set3 stats-set0 stats-set1
Modified: spamassassin/trunk/rulesrc/scores/scores-set3
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/scores-set3?rev=1914758&r1=1914757&r2=1914758&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/scores-set3 (original)
+++ spamassassin/trunk/rulesrc/scores/scores-set3 Tue Dec 19 02:44:31 2023
@@ -1,92 +1,91 @@
# Using score set 1 logs for revision 1914705 from:
# ham-net-ena-week0.r1914705.log ham-net-ena-week1.r1914705.log ham-net-ena-week3.r1914705.log ham-net-ena-week4.r1914705.log ham-net-giovanni-ham.r1914705.log ham-net-giovanni-spammy.r1914705.log ham-net-giovanni-spam.r1914705.log ham-net-grenier.r1914705.log ham-net-hege.r1914705.log ham-net-jhardin.r1914705.log ham-net-llanga.r1914705.log ham-net-mmiroslaw-mails-ham.r1914705.log ham-net-mmiroslaw-mails-spam.r1914705.log ham-net-spamsponge.r1914705.log ham-net-whyscream.r1914705.log spam-net-ena-week0.r1914705.log spam-net-ena-week1.r1914705.log spam-net-ena-week3.r1914705.log spam-net-ena-week4.r1914705.log spam-net-giovanni-ham.r1914705.log spam-net-giovanni-spammy.r1914705.log spam-net-giovanni-spam.r1914705.log spam-net-grenier.r1914705.log spam-net-hege.r1914705.log spam-net-jhardin.r1914705.log spam-net-llanga.r1914705.log spam-net-mmiroslaw-mails-ham.r1914705.log spam-net-mmiroslaw-mails-spam.r1914705.log spam-net-spamsponge.r1914705.log spam-net-whyscream.r1914705.log

-score ACCT_PHISHING_MANY 3.000
+score ACCT_PHISHING_MANY 2.999
score AC_BR_BONANZA 0.001
score AC_DIV_BONANZA 0.001
-score AC_FROM_MANY_DOTS 2.500
score AC_HTML_NONSENSE_TAGS 1.999
-score AC_POST_EXTRAS 1.894
-score ADMITS_SPAM 2.399
+score AC_POST_EXTRAS 2.138
+score ADMITS_SPAM 2.499
score ADVANCE_FEE_2_NEW_MONEY 1.999
score ADVANCE_FEE_3_NEW 3.499
score ADVANCE_FEE_3_NEW_MONEY 0.001
score ADVANCE_FEE_4_NEW 2.199
score ADVANCE_FEE_4_NEW_MONEY 2.399
score ADVANCE_FEE_5_NEW_FRM_MNY 0.001
-score ADVANCE_FEE_5_NEW_MONEY 0.857
-score AD_PREFS 0.303
+score ADVANCE_FEE_5_NEW_MONEY 0.850
+score AD_PREFS 0.039
score AXB_XMAILER_MIMEOLE_OL_024C2 0.001
-score BITCOIN_EXTORT_01 0.002
-score BITCOIN_EXTORT_02 3.453
-score BITCOIN_MALF_HTML 0.001
-score BITCOIN_ONAN 2.999
-score BITCOIN_SPAM_02 2.500
-score BITCOIN_SPAM_05 1.123
-score BITCOIN_SPAM_07 2.032
-score BITCOIN_TOEQFM 1.188
+score BITCOIN_EXTORT_01 0.443
+score BITCOIN_EXTORT_02 2.437
+score BITCOIN_MALF_HTML 0.001 # force non-zero
+score BITCOIN_ONAN 1.988
+score BITCOIN_SPAM_02 2.499
+score BITCOIN_SPAM_05 0.959
+score BITCOIN_SPAM_07 1.243
+score BITCOIN_TOEQFM 1.480
score BITCOIN_VISTA 0.001
-score BITCOIN_XPRIO 1.976
+score BITCOIN_XPRIO 1.974
score BODY_SINGLE_URI 0.001
-score BODY_SINGLE_WORD 0.001 # force non-zero
+score BODY_SINGLE_WORD 0.001
score BODY_URI_ONLY 2.696
-score CK_HELO_GENERIC 0.249
-score CTE_8BIT_MISMATCH 0.001 # force non-zero
-score DEAR_BENEFICIARY 0.399
-score DKIMWL_WL_HIGH -0.083
-score DKIMWL_WL_MED -0.001
-score DKIMWL_WL_MEDHI -0.001 # force non-zero
-score DYNAMIC_IMGUR 3.358
+score CK_HELO_GENERIC 0.250
+score CTE_8BIT_MISMATCH 0.001
+score DEAR_BENEFICIARY 0.434
+score DKIMWL_WL_HIGH -0.066
+score DKIMWL_WL_MED -0.001 # force non-zero
+score DKIMWL_WL_MEDHI -0.001
+score DYNAMIC_IMGUR 2.065
score END_FUTURE_EMAILS 2.499
score FACEBOOK_IMG_NOT_RCVD_FB 1.999
score FILL_THIS_FORM 0.499
-score FONT_INVIS_DIRECT 0.001
+score FONT_INVIS_DIRECT 0.001 # force non-zero
score FONT_INVIS_LONG_LINE 2.999
-score FONT_INVIS_MSGID 2.213
-score FONT_INVIS_NORDNS 1.647
-score FONT_INVIS_POSTEXTRAS 1.918
+score FONT_INVIS_MSGID 1.959
+score FONT_INVIS_NORDNS 0.737
+score FONT_INVIS_POSTEXTRAS 1.676
score FORGED_SPF_HELO 2.999
score FORM_FRAUD 0.999
-score FORM_FRAUD_5 0.526
+score FORM_FRAUD_5 2.422
score FREEMAIL_FORGED_FROMDOMAIN 0.250
score FROM_ADDR_WS 2.999
score FROM_FMBLA_NEWDOM 1.499
score FROM_FMBLA_NEWDOM28 0.799
-score FROM_GOV_DKIM_AU -0.208
-score FROM_IN_TO_AND_SUBJ 1.500
+score FROM_GOV_DKIM_AU -0.999
+score FROM_IN_TO_AND_SUBJ 0.276
score FROM_MISSPACED 1.999
-score FROM_MISSP_DYNIP 0.315
-score FROM_MISSP_EH_MATCH 0.001 # force non-zero
-score FROM_MISSP_FREEMAIL 0.001
-score FROM_MISSP_MSFT 1.649
+score FROM_MISSP_DYNIP 0.331
+score FROM_MISSP_EH_MATCH 0.001
+score FROM_MISSP_FREEMAIL 0.001 # force non-zero
+score FROM_MISSP_MSFT 1.647
score FROM_MISSP_REPLYTO 2.499
-score FROM_MISSP_SPF_FAIL 1.947
-score FROM_MISSP_TO_UNDISC 3.894
+score FROM_MISSP_SPF_FAIL 1.840
+score FROM_MISSP_TO_UNDISC 3.963
score FROM_MISSP_USER 0.001
score FROM_MISSP_XPRIO 2.499
-score FROM_PAYPAL_SPOOF 0.596
+score FROM_PAYPAL_SPOOF 1.599
score FROM_SUSPICIOUS_NTLD 0.499
score FROM_SUSPICIOUS_NTLD_FP 1.999
-score FROM_UNBAL1 2.199
+score FROM_UNBAL1 2.200
score FROM_WSP_TRAIL 2.299
-score FSL_BULK_SIG 1.633
+score FSL_BULK_SIG 1.516
score FSL_CTYPE_WIN1251 0.001
-score FSL_HAS_TINYURL 2.299
+score FSL_HAS_TINYURL 2.399
score FSL_HELO_FAKE 1.199
-score FSL_NEW_HELO_USER 0.001 # force non-zero
+score FSL_NEW_HELO_USER 0.001
score FUZZY_AMAZON 2.499
-score FUZZY_BITCOIN 2.590
+score FUZZY_BITCOIN 2.004
score FUZZY_IMPORTANT 2.399
score FUZZY_WALLET 1.899
-score GB_BITCOIN_CP 1.029
-score GB_CUSTOM_HTM_URI 1.338
-score GB_FAKE_RF_SHORT 1.105
+score GB_BITCOIN_CP 1.085
+score GB_CUSTOM_HTM_URI 1.042
+score GB_FAKE_RF_SHORT 0.968
score GB_FREEMAIL_DISPTO 0.001
score GB_HASHBL_BTC 0.001
-score GOOG_REDIR_NORDNS 0.274
+score GOOG_REDIR_NORDNS 0.275
score GOOG_STO_IMG_HTML 2.723
score GOOG_STO_NOIMG_HTML 3.000
-score HAS_X_OUTGOING_SPAM_STAT 0.001
+score HAS_X_OUTGOING_SPAM_STAT 0.001 # force non-zero
score HDRS_LCASE_IMGONLY 0.100
score HDRS_MISSP 2.499
score HDR_ORDER_FTSDMCXX_DIRECT 0.001
@@ -98,111 +97,113 @@ score HK_NAME_FM_MR_MRS 1.4
score HK_NAME_MR_MRS 0.999
score HK_RANDOM_ENVFROM 0.999
score HK_RANDOM_FROM 0.999
-score HK_RANDOM_REPLYTO 1.000
+score HK_RANDOM_REPLYTO 0.999
score HK_SCAM 1.999
-score HOSTED_IMG_DIRECT_MX 3.139
-score HOSTED_IMG_FREEM 2.428
+score HOSTED_IMG_DIRECT_MX 3.083
+score HOSTED_IMG_FREEM 2.432
score HOSTED_IMG_MULTI_PUB_01 2.999
-score HTML_BADATTR 1.000
+score HTML_BADATTR 0.999
score HTML_FONT_TINY_NORDNS 0.001
-score HTML_SINGLET_MANY 2.500
-score HTML_TEXT_INVISIBLE_FONT 1.439
-score HTML_TEXT_INVISIBLE_STYLE 3.500
+score HTML_SINGLET_MANY 2.499
+score HTML_TEXT_INVISIBLE_FONT 1.694
+score HTML_TEXT_INVISIBLE_STYLE 3.499
score JH_SPAMMY_HEADERS 3.499
-score KHOP_HELO_FCRDNS 0.261
+score KHOP_HELO_FCRDNS 0.260
score LIST_PARTIAL_SHORT_MSG 2.170
score LONG_HEX_URI 2.999
-score LONG_IMG_URI 3.000
-score LONG_INVISIBLE_TEXT 2.720
+score LONG_IMG_URI 2.999
+score LONG_INVISIBLE_TEXT 2.857
score LOTS_OF_MONEY 0.010
-score LOTTO_DEPT 2.000
-score MALFORMED_FREEMAIL 2.199
-score MALWARE_NORDNS 3.337
+score LOTTO_DEPT 1.999
+score MALFORMED_FREEMAIL 2.300
+score MALWARE_NORDNS 0.907
score MILLION_HUNDRED 0.001
-score MILLION_USD 1.089
+score MILLION_USD 1.155
score MIMEOLE_DIRECT_TO_MX 0.001
-score MIME_NO_TEXT 0.001
+score MIME_NO_TEXT 1.575
score MIXED_ES 2.099
score MIXED_HREF_CASE 1.999
-score MONEY_ATM_CARD 1.102
-score MONEY_FORM 0.001
-score MONEY_FORM_SHORT 0.001
-score MONEY_FRAUD_3 0.247
-score MONEY_FRAUD_5 2.507
-score MONEY_FRAUD_8 0.666
-score MONEY_FREEMAIL_REPTO 1.402
+score MONEY_ATM_CARD 0.917
+score MONEY_FORM 0.001 # force non-zero
+score MONEY_FORM_SHORT 0.001 # force non-zero
+score MONEY_FRAUD_3 0.078
+score MONEY_FRAUD_5 2.559
+score MONEY_FRAUD_8 0.506
+score MONEY_FREEMAIL_REPTO 1.407
score MONEY_FROM_MISSP 0.001
-score MONEY_NOHTML 1.489
+score MONEY_NOHTML 1.435
score NA_DOLLARS 1.499
-score NICE_REPLY_A -2.342
-score NO_FM_NAME_IP_HOSTN 0.001 # force non-zero
+score NICE_REPLY_A -2.345
+score NO_FM_NAME_IP_HOSTN 0.001
score NSL_RCVD_FROM_USER 0.001
-score NSL_RCVD_HELO_USER 0.095
-score OBFU_BITCOIN 1.131
+score NSL_RCVD_HELO_USER 0.174
+score OBFU_BITCOIN 1.712
score OBFU_TEXT_ATTACH 1.599
score PDS_BAD_THREAD_QP_64 0.999
score PDS_BTC_ID 0.001 # force non-zero
score PDS_BTC_MSGID 0.001
score PDS_EMPTYSUBJ_URISHRT 1.499
score PDS_FRNOM_TODOM_DBL_URL 0.001
-score PDS_FRNOM_TODOM_NAKED_TO 1.499
-score PDS_FROM_NAME_TO_DOMAIN 2.000
+score PDS_FRNOM_TODOM_NAKED_TO 1.500
+score PDS_FROM_NAME_TO_DOMAIN 1.999
score PDS_HELO_SPF_FAIL 1.999
-score PDS_HP_HELO_NORDNS 0.001 # force non-zero
+score PDS_HP_HELO_NORDNS 0.001
score PDS_NAKED_TO_NUMERO 1.999
score PDS_OTHER_BAD_TLD 1.999
score PDS_TINYSUBJ_URISHRT 1.499
score PHP_SCRIPT 2.398
score PP_MIME_FAKE_ASCII_TEXT 0.001
score RAND_MKTG_HEADER 1.999
-score RATWARE_NO_RDNS 2.124
+score RATWARE_NO_RDNS 2.266
score RCVD_IN_MSPIKE_H2 -0.001
score SCC_BODY_SINGLE_WORD 0.001
score SCC_CANSPAM_2 2.899
score SCC_ISEMM_LID_1B 1.499
-score SENDGRID_REDIR 0.534
+score SENDGRID_REDIR 0.674
score SHOPIFY_IMG_NOT_RCVD_SFY 2.499
-score SHORTENED_URL_SRC 1.544
-score SHORT_SHORTNER 1.999
+score SHORTENED_URL_SRC 1.552
+score SHORT_SHORTNER 2.000
score SPOOFED_FREEMAIL 0.001
score SPOOFED_FREEMAIL_NO_RDNS 0.001
score SPOOFED_FREEM_REPTO 0.001
-score STATIC_XPRIO_OLE 1.266
+score STATIC_XPRIO_OLE 1.268
score SUBJ_ATTENTION 0.499
score THIS_AD 0.799
score TONOM_EQ_TOLOC_SHRT_SHRTNER 1.499
-score TO_EQ_FM_DIRECT_MX 2.499
-score TO_EQ_FM_DOM_SPF_FAIL 0.902
-score TO_EQ_FM_SPF_FAIL 0.001
+score TO_EQ_FM_DIRECT_MX 1.758
+score TO_EQ_FM_DOM_HTML_ONLY 1.408
+score TO_EQ_FM_DOM_SPF_FAIL 0.001
+score TO_EQ_FM_SPF_FAIL 0.335
score TO_IN_SUBJ 0.099
score TO_NO_BRKTS_FROM_MSSP 2.499
-score TO_NO_BRKTS_HTML_IMG 1.877
+score TO_NO_BRKTS_HTML_IMG 1.819
score TO_NO_BRKTS_HTML_ONLY 1.999
score TO_NO_BRKTS_MSFT 2.499
score TO_NO_BRKTS_NORDNS_HTML 1.999
-score TO_NO_BRKTS_PCNT 2.034
+score TO_NO_BRKTS_PCNT 2.139
score TVD_SPACE_RATIO_MINFP 0.001
score UNDISC_FREEM 2.599
-score UNDISC_MONEY 2.799
+score UNDISC_MONEY 2.899
score UNICODE_OBFU_ZW_MANY 2.999
-score URI_DOTEDU 1.543
+score URI_DOTEDU 1.837
score URI_GOOGLE_PROXY 1.699
score URI_IN_URI_5 3.199
score URI_PHISH 3.999
-score URI_PHP_REDIR 2.265
+score URI_PHP_REDIR 2.263
score URI_TRY_3LD 1.999
score URI_WP_HACKED_2 2.499
score VFY_ACCT_NORDNS 2.999
score VISTA_COST 2.499
-score VISTA_TONOM_EQ_TOLOC 2.500
-score WIKI_IMG 1.999
-score WORD_INVIS 1.892
+score VISTA_TONOM_EQ_TOLOC 2.499
+score WIKI_IMG 2.099
+score WORD_INVIS 1.899
score WORD_INVIS_MANY 2.999
score XFER_LOTSA_MONEY 0.001
-score XM_DIGITS_ONLY 1.849
+score XM_DIGITS_ONLY 1.507
score XM_RANDOM 2.499
-score XM_UC_ONLY 2.500
+score XM_UC_ONLY 2.499
score XPRIO 0.001
+score AC_FROM_MANY_DOTS 1.000
score AC_SPAMMY_URI_PATTERNS1 1.000
score AC_SPAMMY_URI_PATTERNS10 1.000
score AC_SPAMMY_URI_PATTERNS11 1.000

Modified: spamassassin/trunk/rulesrc/scores/stats-set0
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set0?rev=1914758&r1=1914757&r2=1914758&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set0 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set0 Tue Dec 19 02:44:31 2023
@@ -1,40 +1,40 @@
##### WITH NEW RULES AND SCORES #####

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 459036 62.076% (99.212% of non-spam corpus)
-# Correctly spam: 173200 23.422% (62.573% of spam corpus)
-# False positives: 3646 0.493% (0.788% of nonspam, 72721 weighted)
-# False negatives: 103596 14.009% (37.427% of spam, 212707 weighted)
-# Average score for spam: 7.0 nonspam: 0.5
-# Average for false-pos: 6.7 false-neg: 2.1
-# TOTAL: 739478 100.00%
+# Correctly non-spam: 461686 61.503% (99.166% of non-spam corpus)
+# Correctly spam: 181099 24.125% (63.519% of spam corpus)
+# False positives: 3881 0.517% (0.834% of nonspam, 77983 weighted)
+# False negatives: 104012 13.856% (36.481% of spam, 210964 weighted)
+# Average score for spam: 7.1 nonspam: 0.6
+# Average for false-pos: 6.6 false-neg: 2.0
+# TOTAL: 750678 100.00%

Reading scores from "tmprules"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 57224 99.19%
-# Correctly spam: 21484 62.05%
-# False positives: 468 0.81%
-# False negatives: 13137 37.95%
-# TCR(l=50): 0.947560 SpamRecall: 62.055% SpamPrec: 97.868%
+# Correctly non-spam: 57634 99.24%
+# Correctly spam: 22167 62.18%
+# False positives: 440 0.76%
+# False negatives: 13483 37.82%
+# TCR(l=50): 1.004706 SpamRecall: 62.180% SpamPrec: 98.054%

##### WITHOUT NEW RULES AND SCORES #####
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 450057 97.27%
-# Correctly spam: 113941 41.16%
-# False positives: 12625 2.73%
-# False negatives: 162855 58.84%
-# TCR(l=50): 0.348563 SpamRecall: 41.164% SpamPrec: 90.025%
+# Correctly non-spam: 452875 97.27%
+# Correctly spam: 119140 41.79%
+# False positives: 12692 2.73%
+# False negatives: 165971 58.21%
+# TCR(l=50): 0.356135 SpamRecall: 41.787% SpamPrec: 90.373%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 56145 97.32%
-# Correctly spam: 14112 40.76%
-# False positives: 1547 2.68%
-# False negatives: 20509 59.24%
-# TCR(l=50): 0.353785 SpamRecall: 40.761% SpamPrec: 90.121%
+# Correctly non-spam: 56469 97.24%
+# Correctly spam: 14730 41.32%
+# False positives: 1605 2.76%
+# False negatives: 20920 58.68%
+# TCR(l=50): 0.352377 SpamRecall: 41.318% SpamPrec: 90.174%

Modified: spamassassin/trunk/rulesrc/scores/stats-set1
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set1?rev=1914758&r1=1914757&r2=1914758&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set1 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set1 Tue Dec 19 02:44:31 2023
@@ -1,23 +1,23 @@
##### WITH NEW RULES AND SCORES #####

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 406713 84.541% (99.241% of non-spam corpus)
-# Correctly spam: 30291 6.296% (42.507% of spam corpus)
-# False positives: 3109 0.646% (0.759% of nonspam, 33743 weighted)
-# False negatives: 40971 8.516% (57.493% of spam, 67311 weighted)
+# Correctly non-spam: 406704 84.539% (99.239% of non-spam corpus)
+# Correctly spam: 30341 6.307% (42.577% of spam corpus)
+# False positives: 3118 0.648% (0.761% of nonspam, 33843 weighted)
+# False negatives: 40920 8.506% (57.423% of spam, 67192 weighted)
# Average score for spam: 6.0 nonspam: -1.0
# Average for false-pos: 6.2 false-neg: 1.6
-# TOTAL: 481084 100.00%
+# TOTAL: 481083 100.00%

Reading scores from "tmprules"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 50761 99.33%
-# Correctly spam: 3841 42.75%
-# False positives: 341 0.67%
-# False negatives: 5143 57.25%
-# TCR(l=50): 0.404812 SpamRecall: 42.754% SpamPrec: 91.846%
+# Correctly non-spam: 50762 99.33%
+# Correctly spam: 3846 42.81%
+# False positives: 340 0.67%
+# False negatives: 5138 57.19%
+# TCR(l=50): 0.405818 SpamRecall: 42.809% SpamPrec: 91.878%

##### WITHOUT NEW RULES AND SCORES #####
Reading scores from "../rules-base"...
@@ -25,16 +25,16 @@ Reading per-message hit stat logs and sc

# SUMMARY for threshold 5.0:
# Correctly non-spam: 408145 99.59%
-# Correctly spam: 14700 20.63%
+# Correctly spam: 14685 20.61%
# False positives: 1677 0.41%
-# False negatives: 56562 79.37%
-# TCR(l=50): 0.507521 SpamRecall: 20.628% SpamPrec: 89.760%
+# False negatives: 56576 79.39%
+# TCR(l=50): 0.507463 SpamRecall: 20.607% SpamPrec: 89.751%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
# Correctly non-spam: 50903 99.61%
-# Correctly spam: 1838 20.46%
+# Correctly spam: 1858 20.68%
# False positives: 199 0.39%
-# False negatives: 7146 79.54%
-# TCR(l=50): 0.525503 SpamRecall: 20.459% SpamPrec: 90.231%
+# False negatives: 7126 79.32%
+# TCR(l=50): 0.526119 SpamRecall: 20.681% SpamPrec: 90.326%