Mailing List Archive

svn commit: r1914983 [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=1914983&r1=1914982&r2=1914983&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/scores-set3 (original)
+++ spamassassin/trunk/rulesrc/scores/scores-set3 Sat Dec 30 02:42:09 2023
@@ -1,92 +1,90 @@
# Using score set 1 logs for revision 1914880 from:
# ham-net-ena-week0.r1914880.log ham-net-ena-week1.r1914880.log ham-net-ena-week2.r1914880.log ham-net-ena-week3.r1914880.log ham-net-ena-week4.r1914880.log ham-net-giovanni-ham.r1914880.log ham-net-giovanni-spammy.r1914880.log ham-net-giovanni-spam.r1914880.log ham-net-grenier.r1914880.log ham-net-hege.r1914880.log ham-net-jhardin.r1914880.log ham-net-llanga.r1914880.log ham-net-mmiroslaw-mails-ham.r1914880.log ham-net-mmiroslaw-mails-spam.r1914880.log ham-net-spamsponge.r1914880.log ham-net-whyscream.r1914880.log spam-net-ena-week0.r1914880.log spam-net-ena-week1.r1914880.log spam-net-ena-week2.r1914880.log spam-net-ena-week3.r1914880.log spam-net-ena-week4.r1914880.log spam-net-giovanni-ham.r1914880.log spam-net-giovanni-spammy.r1914880.log spam-net-giovanni-spam.r1914880.log spam-net-grenier.r1914880.log spam-net-hege.r1914880.log spam-net-jhardin.r1914880.log spam-net-llanga.r1914880.log spam-net-mmiroslaw-mails-ham.r1914880.log spam-net-mmiroslaw-mails-spam.r1914880.log spam-n
et-spamsponge.r1914880.log spam-net-whyscream.r1914880.log

-score ACCT_PHISHING_MANY 2.997
+score ACCT_PHISHING_MANY 2.996
score AC_BR_BONANZA 0.001
score AC_DIV_BONANZA 0.001
-score AC_HTML_NONSENSE_TAGS 1.997
-score AC_POST_EXTRAS 1.645
+score AC_HTML_NONSENSE_TAGS 1.998
+score AC_POST_EXTRAS 1.784
score ADMITS_SPAM 1.697
score ADVANCE_FEE_2_NEW_MONEY 1.997
-score ADVANCE_FEE_3_NEW 3.266
+score ADVANCE_FEE_3_NEW 3.273
score ADVANCE_FEE_3_NEW_MONEY 0.001
-score ADVANCE_FEE_4_NEW 2.197
score ADVANCE_FEE_4_NEW_MONEY 0.684
score ADVANCE_FEE_5_NEW_FRM_MNY 0.001
-score ADVANCE_FEE_5_NEW_MONEY 1.685
-score AD_PREFS 0.402
+score ADVANCE_FEE_5_NEW_MONEY 1.109
+score AD_PREFS 0.011
score AXB_XMAILER_MIMEOLE_OL_024C2 0.001
-score BITCOIN_EXTORT_01 2.005
-score BITCOIN_MALF_HTML 2.765
-score BITCOIN_ONAN 2.511
+score BITCOIN_EXTORT_01 2.013
+score BITCOIN_MALF_HTML 3.082
+score BITCOIN_ONAN 2.893
score BITCOIN_SPAM_02 2.497
-score BITCOIN_SPAM_05 1.079
-score BITCOIN_SPAM_07 0.002
-score BITCOIN_TOEQFM 2.627
+score BITCOIN_SPAM_05 1.447
+score BITCOIN_SPAM_07 0.125
+score BITCOIN_TOEQFM 2.494
score BITCOIN_VISTA 0.001
-score BITCOIN_XPRIO 1.654
-score BITCOIN_YOUR_INFO 2.705
+score BITCOIN_XPRIO 1.468
+score BITCOIN_YOUR_INFO 2.693
score BODY_SINGLE_URI 0.001
score BODY_SINGLE_WORD 0.001
-score BODY_URI_ONLY 0.206
-score CK_HELO_GENERIC 0.249
-score CTE_8BIT_MISMATCH 0.001 # force non-zero
+score BODY_URI_ONLY 0.181
+score CK_HELO_GENERIC 0.248
+score CTE_8BIT_MISMATCH 0.001
score DEAR_BENEFICIARY 1.864
-score DKIMWL_WL_HIGH -2.553
-score DKIMWL_WL_MED -0.001 # force non-zero
+score DKIMWL_WL_HIGH -2.667
+score DKIMWL_WL_MED -0.001
score DKIMWL_WL_MEDHI -0.001
-score DYNAMIC_IMGUR 3.167
-score END_FUTURE_EMAILS 2.497
-score FILL_THIS_FORM 0.455
+score DYNAMIC_IMGUR 2.618
+score END_FUTURE_EMAILS 2.498
+score FILL_THIS_FORM 0.451
score FONT_INVIS_DIRECT 0.001
score FONT_INVIS_LONG_LINE 2.996
-score FONT_INVIS_MSGID 1.184
-score FONT_INVIS_NORDNS 1.606
-score FONT_INVIS_POSTEXTRAS 2.161
-score FORGED_SPF_HELO 2.998
+score FONT_INVIS_MSGID 1.245
+score FONT_INVIS_NORDNS 1.451
+score FONT_INVIS_POSTEXTRAS 2.026
+score FORGED_SPF_HELO 3.014
score FORM_FRAUD 0.998
-score FORM_FRAUD_5 1.154
-score FREEMAIL_FORGED_FROMDOMAIN 0.248
+score FORM_FRAUD_5 1.362
+score FREEMAIL_FORGED_FROMDOMAIN 0.249
score FROM_ADDR_WS 2.996
score FROM_FMBLA_NEWDOM14 0.998
score FROM_FMBLA_NEWDOM28 0.798
-score FROM_GOV_DKIM_AU -0.999
+score FROM_GOV_DKIM_AU -0.930
score FROM_IN_TO_AND_SUBJ 1.498
score FROM_MISSPACED 1.997
score FROM_MISSP_DYNIP 0.001
score FROM_MISSP_EH_MATCH 0.001
score FROM_MISSP_FREEMAIL 0.001
score FROM_MISSP_MSFT 1.349
-score FROM_MISSP_REPLYTO 1.638
-score FROM_MISSP_SPF_FAIL 1.999
-score FROM_MISSP_TO_UNDISC 3.622
+score FROM_MISSP_REPLYTO 1.660
+score FROM_MISSP_SPF_FAIL 1.499
+score FROM_MISSP_TO_UNDISC 3.360
score FROM_MISSP_USER 0.001
-score FROM_MISSP_XPRIO 0.944
+score FROM_MISSP_XPRIO 0.942
score FROM_PAYPAL_SPOOF 0.001
-score FROM_SUSPICIOUS_NTLD 0.498
+score FROM_SUSPICIOUS_NTLD 0.416
score FROM_SUSPICIOUS_NTLD_FP 1.997
-score FSL_BULK_SIG 0.001 # force non-zero
+score FSL_BULK_SIG 0.001
score FSL_CTYPE_WIN1251 0.001
-score FSL_HAS_TINYURL 2.397
score FSL_NEW_HELO_USER 0.001
-score FUZZY_AMAZON 2.397
-score FUZZY_BITCOIN 1.157
-score FUZZY_IMPORTANT 1.549
+score FUZZY_AMAZON 2.497
+score FUZZY_BITCOIN 0.997
+score FUZZY_IMPORTANT 1.027
score FUZZY_WALLET 1.797
-score GB_BITCOIN_CP 1.849
-score GB_CUSTOM_HTM_URI 1.498
-score GB_FAKE_RF_SHORT 1.112
+score GB_BITCOIN_CP 2.802
+score GB_CUSTOM_HTM_URI 0.830
+score GB_FAKE_RF_SHORT 1.128
score GB_FREEMAIL_DISPTO 0.001
-score GB_HASHBL_BTC 0.393
+score GB_HASHBL_BTC 0.001
score GOOG_REDIR_NORDNS 0.001
-score GOOG_STO_IMG_HTML 2.522
+score GOOG_STO_IMG_HTML 2.535
score GOOG_STO_NOIMG_HTML 2.996
score HDRS_LCASE_IMGONLY 0.099
score HDRS_MISSP 2.497
score HDR_ORDER_FTSDMCXX_DIRECT 0.001
score HDR_ORDER_FTSDMCXX_NORDNS 0.001
score HEADER_FROM_DIFFERENT_DOMAINS 0.249
-score HELO_MISC_IP 0.248
+score HELO_MISC_IP 0.249
score HELO_NO_DOMAIN 0.001
score HK_NAME_FM_MR_MRS 1.498
score HK_NAME_MR_MRS 0.998
@@ -95,106 +93,107 @@ score HK_RANDOM_FROM 0.9
score HK_RANDOM_REPLYTO 0.998
score HK_SCAM 1.997
score HOSTED_IMG_DIRECT_MX 3.496
-score HOSTED_IMG_FREEM 3.209
+score HOSTED_IMG_FREEM 3.201
score HOSTED_IMG_MULTI_PUB_01 2.996
score HTML_BADATTR 0.998
score HTML_FONT_TINY_NORDNS 0.001
-score HTML_SINGLET_MANY 2.498
-score HTML_TEXT_INVISIBLE_FONT 1.363
+score HTML_SINGLET_MANY 2.497
+score HTML_TEXT_INVISIBLE_FONT 1.308
score HTML_TEXT_INVISIBLE_STYLE 2.426
-score JH_SPAMMY_HEADERS 3.497
-score KHOP_HELO_FCRDNS 0.259
+score JH_SPAMMY_HEADERS 3.496
+score KHOP_HELO_FCRDNS 0.258
score LONG_HEX_URI 2.996
score LONG_IMG_URI 2.996
-score LONG_INVISIBLE_TEXT 2.696
+score LONG_INVISIBLE_TEXT 2.693
score LOTS_OF_MONEY 0.010
score LOTTO_DEPT 1.997
score MALFORMED_FREEMAIL 2.397
score MILLION_HUNDRED 0.001
score MILLION_USD 1.997
score MIMEOLE_DIRECT_TO_MX 0.001
-score MIXED_CENTER_CASE 2.219
-score MIXED_ES 2.097
+score MIXED_CENTER_CASE 2.006
+score MIXED_ES 2.197
score MIXED_HREF_CASE 1.997
-score MONEY_ATM_CARD 1.417
+score MONEY_ATM_CARD 1.104
score MONEY_FORM 0.001
-score MONEY_FORM_SHORT 2.497
+score MONEY_FORM_SHORT 2.498
score MONEY_FRAUD_3 0.001
-score MONEY_FRAUD_5 2.596
-score MONEY_FRAUD_8 0.325
-score MONEY_FREEMAIL_REPTO 1.389
+score MONEY_FRAUD_5 2.597
+score MONEY_FRAUD_8 1.235
+score MONEY_FREEMAIL_REPTO 1.386
score MONEY_FROM_MISSP 0.001
score NA_DOLLARS 1.498
-score NICE_REPLY_A -2.685
+score NICE_REPLY_A -2.335
score NO_FM_NAME_IP_HOSTN 0.001 # force non-zero
score NSL_RCVD_FROM_USER 0.001
-score NSL_RCVD_HELO_USER 0.289
-score OBFU_BITCOIN 1.488
+score NSL_RCVD_HELO_USER 0.001
+score OBFU_BITCOIN 1.262
score OBFU_TEXT_ATTACH 1.597
-score ODD_FREEM_REPTO 2.254
+score ODD_FREEM_REPTO 2.247
score PDS_BAD_THREAD_QP_64 0.998
-score PDS_BTC_ID 0.001
-score PDS_BTC_MSGID 0.001 # force non-zero
+score PDS_BTC_ID 0.386
+score PDS_BTC_MSGID 0.001
score PDS_EMPTYSUBJ_URISHRT 1.498
score PDS_FRNOM_TODOM_DBL_URL 0.001
score PDS_FRNOM_TODOM_NAKED_TO 1.498
+score PDS_FROM_2_EMAILS 1.682
score PDS_FROM_NAME_TO_DOMAIN 1.997
score PDS_HELO_SPF_FAIL 1.997
score PDS_HP_HELO_NORDNS 0.001
-score PDS_NAKED_TO_NUMERO 1.998
+score PDS_NAKED_TO_NUMERO 1.997
score PDS_TINYSUBJ_URISHRT 1.498
-score PHP_SCRIPT 2.303
+score PHP_SCRIPT 2.298
score PP_MIME_FAKE_ASCII_TEXT 0.001
score RAND_MKTG_HEADER 1.997
-score RATWARE_NO_RDNS 2.906
+score RATWARE_NO_RDNS 1.182
score RCVD_IN_MSPIKE_H2 -0.001
score SCC_BODY_SINGLE_WORD 0.001
-score SCC_BODY_URI_ONLY 2.061
-score SCC_CANSPAM_2 2.997
+score SCC_BODY_URI_ONLY 2.081
+score SCC_CANSPAM_2 2.996
score SCC_ISEMM_LID_1B 1.498
-score SCC_NEWBIE_HASBEENS 2.109
-score SCC_SPAMMER_ADDR_2 0.871
-score SENDGRID_REDIR 1.066
+score SCC_NEWBIE_HASBEENS 2.198
+score SCC_SPAMMER_ADDR_2 0.974
+score SENDGRID_REDIR 1.067
score SHOPIFY_IMG_NOT_RCVD_SFY 2.497
score SHORTENED_URL_SRC 2.397
score SHORT_SHORTNER 1.997
score SPOOFED_FREEMAIL 0.001
score SPOOFED_FREEMAIL_NO_RDNS 0.001
-score SPOOFED_FREEM_REPTO 1.368
-score STATIC_XPRIO_OLE 0.766
+score SPOOFED_FREEM_REPTO 1.539
+score STATIC_XPRIO_OLE 0.776
score SUBJ_ATTENTION 0.498
score THIS_AD 0.698
score TO_EQ_FM_DIRECT_MX 2.497
-score TO_EQ_FM_DOM_SPF_FAIL 0.961
-score TO_EQ_FM_SPF_FAIL 0.780
-score TO_IN_SUBJ 0.098
-score TO_NAME_SUBJ_NO_RDNS 2.894
+score TO_EQ_FM_DOM_SPF_FAIL 0.824
+score TO_EQ_FM_SPF_FAIL 0.918
+score TO_IN_SUBJ 0.099
+score TO_NAME_SUBJ_NO_RDNS 2.905
score TO_NO_BRKTS_FROM_MSSP 2.497
-score TO_NO_BRKTS_HTML_IMG 1.998
+score TO_NO_BRKTS_HTML_IMG 1.997
score TO_NO_BRKTS_HTML_ONLY 1.997
-score TO_NO_BRKTS_MSFT 2.272
+score TO_NO_BRKTS_MSFT 2.273
score TO_NO_BRKTS_NORDNS_HTML 1.997
score TO_NO_BRKTS_PCNT 2.499
score TVD_SPACE_RATIO_MINFP 0.001
score UNDISC_FREEM 2.596
-score UNDISC_MONEY 2.796
-score UNICODE_OBFU_ZW_MANY 0.484
+score UNDISC_MONEY 2.896
+score UNICODE_OBFU_ZW_MANY 0.499
score URI_DOTEDU 1.997
score URI_GOOGLE_PROXY 1.597
score URI_PHISH 3.995
-score URI_TRY_3LD 1.998
+score URI_TRY_3LD 1.997
score URI_WP_HACKED 3.496
-score URI_WP_HACKED_2 2.498
+score URI_WP_HACKED_2 2.497
score VFY_ACCT_NORDNS 2.996
score VISTA_TONOM_EQ_TOLOC 2.497
score WIKI_IMG 1.997
-score WORD_INVIS 1.897
-score WORD_INVIS_MANY 2.997
+score WORD_INVIS 1.892
+score WORD_INVIS_MANY 2.996
score XFER_LOTSA_MONEY 0.001
-score XM_DIGITS_ONLY 2.585
+score XM_DIGITS_ONLY 0.446
score XM_RANDOM 2.497
score XM_UC_ONLY 2.497
-score XPRIO 0.001
+score XPRIO 0.001 # force non-zero
score AC_FROM_MANY_DOTS 1.000
score AC_SPAMMY_URI_PATTERNS1 1.000
score AC_SPAMMY_URI_PATTERNS10 1.000
@@ -306,6 +305,7 @@ score LUCRATIVE 1.0
score MALF_HTML_B64 1.000
score MALWARE_NORDNS 1.000
score MALWARE_PASSWORD 1.000
+score MANY_HDRS_LCASE 0.100
score MIME_NO_TEXT 1.000
score MIXED_AREA_CASE 1.000
score MIXED_FONT_CASE 1.000

Modified: spamassassin/trunk/rulesrc/scores/stats-set0
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set0?rev=1914983&r1=1914982&r2=1914983&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set0 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set0 Sat Dec 30 02:42:09 2023
@@ -1,40 +1,40 @@
##### WITH NEW RULES AND SCORES #####

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 457760 61.092% (99.198% of non-spam corpus)
-# Correctly spam: 197462 26.353% (68.603% of spam corpus)
-# False positives: 3700 0.494% (0.802% of nonspam, 74836 weighted)
-# False negatives: 90372 12.061% (31.397% of spam, 166772 weighted)
-# Average score for spam: 6.8 nonspam: 0.6
-# Average for false-pos: 6.5 false-neg: 1.8
-# TOTAL: 749294 100.00%
+# Correctly non-spam: 457426 61.387% (99.207% of non-spam corpus)
+# Correctly spam: 194106 26.049% (68.332% of spam corpus)
+# False positives: 3656 0.491% (0.793% of nonspam, 72720 weighted)
+# False negatives: 89958 12.073% (31.668% of spam, 166805 weighted)
+# Average score for spam: 6.9 nonspam: 0.6
+# Average for false-pos: 6.5 false-neg: 1.9
+# TOTAL: 745146 100.00%

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

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 57084 99.21%
-# Correctly spam: 23588 65.56%
-# False positives: 452 0.79%
-# False negatives: 12393 34.44%
-# TCR(l=50): 1.028234 SpamRecall: 65.557% SpamPrec: 98.120%
+# Correctly non-spam: 56976 99.10%
+# Correctly spam: 23055 64.89%
+# False positives: 515 0.90%
+# False negatives: 12473 35.11%
+# TCR(l=50): 0.929493 SpamRecall: 64.892% SpamPrec: 97.815%

##### 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: 449270 97.36%
-# Correctly spam: 129364 44.94%
-# False positives: 12190 2.64%
-# False negatives: 158470 55.06%
-# TCR(l=50): 0.374798 SpamRecall: 44.944% SpamPrec: 91.388%
+# Correctly non-spam: 448855 97.35%
+# Correctly spam: 129485 45.58%
+# False positives: 12227 2.65%
+# False negatives: 154579 54.42%
+# TCR(l=50): 0.370875 SpamRecall: 45.583% SpamPrec: 91.372%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 56048 97.41%
-# Correctly spam: 16288 45.27%
-# False positives: 1488 2.59%
-# False negatives: 19693 54.73%
-# TCR(l=50): 0.382398 SpamRecall: 45.268% SpamPrec: 91.629%
+# Correctly non-spam: 55969 97.35%
+# Correctly spam: 16170 45.51%
+# False positives: 1522 2.65%
+# False negatives: 19358 54.49%
+# TCR(l=50): 0.372185 SpamRecall: 45.513% SpamPrec: 91.397%

Modified: spamassassin/trunk/rulesrc/scores/stats-set1
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set1?rev=1914983&r1=1914982&r2=1914983&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set1 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set1 Sat Dec 30 02:42:09 2023
@@ -1,23 +1,23 @@
##### WITH NEW RULES AND SCORES #####

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 443129 82.689% (99.366% of non-spam corpus)
-# Correctly spam: 47510 8.865% (52.824% of spam corpus)
-# False positives: 2828 0.528% (0.634% of nonspam, 36799 weighted)
-# False negatives: 42431 7.918% (47.176% of spam, 64669 weighted)
+# Correctly non-spam: 443117 82.687% (99.363% of non-spam corpus)
+# Correctly spam: 47647 8.891% (52.976% of spam corpus)
+# False positives: 2840 0.530% (0.637% of nonspam, 36919 weighted)
+# False negatives: 42293 7.892% (47.024% of spam, 63805 weighted)
# Average score for spam: 6.4 nonspam: -1.6
-# Average for false-pos: 6.5 false-neg: 1.5
-# TOTAL: 535898 100.00%
+# Average for false-pos: 6.4 false-neg: 1.5
+# TOTAL: 535897 100.00%

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

# SUMMARY for threshold 5.0:
# Correctly non-spam: 55265 99.38%
-# Correctly spam: 5944 52.28%
+# Correctly spam: 5997 52.74%
# False positives: 344 0.62%
-# False negatives: 5426 47.72%
-# TCR(l=50): 0.502519 SpamRecall: 52.278% SpamPrec: 94.529%
+# False negatives: 5373 47.26%
+# TCR(l=50): 0.503699 SpamRecall: 52.744% SpamPrec: 94.575%

##### 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: 444141 99.59%
-# Correctly spam: 19550 21.74%
+# Correctly spam: 19557 21.74%
# False positives: 1816 0.41%
-# False negatives: 70391 78.26%
-# TCR(l=50): 0.557978 SpamRecall: 21.736% SpamPrec: 91.501%
+# False negatives: 70383 78.26%
+# TCR(l=50): 0.557999 SpamRecall: 21.744% SpamPrec: 91.503%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
# Correctly non-spam: 55381 99.59%
-# Correctly spam: 2476 21.78%
+# Correctly spam: 2477 21.79%
# False positives: 228 0.41%
-# False negatives: 8894 78.22%
-# TCR(l=50): 0.560264 SpamRecall: 21.777% SpamPrec: 91.568%
+# False negatives: 8893 78.21%
+# TCR(l=50): 0.560292 SpamRecall: 21.785% SpamPrec: 91.571%