You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@spamassassin.apache.org by sp...@apache.org on 2023/08/15 02:42:31 UTC
svn commit: r1911680 [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=1911680&r1=1911679&r2=1911680&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/scores-set3 (original)
+++ spamassassin/trunk/rulesrc/scores/scores-set3 Tue Aug 15 02:42:31 2023
@@ -2,192 +2,189 @@
# ham-net-ena-week0.r1911609.log ham-net-ena-week1.r1911609.log ham-net-ena-week3.r1911609.log ham-net-ena-week4.r1911609.log ham-net-giovanni-ham.r1911609.log ham-net-giovanni-spammy.r1911609.log ham-net-giovanni-spam.r1911609.log ham-net-grenier.r1911609.log ham-net-hege.r1911609.log ham-net-jbrooks.r1911609.log ham-net-jhardin.r1911609.log ham-net-llanga.r1911609.log ham-net-mmiroslaw-mails-ham.r1911609.log ham-net-mmiroslaw-mails-spam.r1911609.log ham-net-spamsponge.r1911609.log ham-net-whyscream.r1911609.log spam-net-ena-week0.r1911609.log spam-net-ena-week1.r1911609.log spam-net-ena-week3.r1911609.log spam-net-ena-week4.r1911609.log spam-net-giovanni-ham.r1911609.log spam-net-giovanni-spammy.r1911609.log spam-net-giovanni-spam.r1911609.log spam-net-grenier.r1911609.log spam-net-hege.r1911609.log spam-net-jbrooks.r1911609.log spam-net-jhardin.r1911609.log spam-net-llanga.r1911609.log spam-net-mmiroslaw-mails-ham.r1911609.log spam-net-mmiroslaw-mails-spam.r1911609.log spam-net-s
pamsponge.r1911609.log spam-net-whyscream.r1911609.log
score AC_DIV_BONANZA 0.001
-score AC_FROM_MANY_DOTS 2.999
+score AC_FROM_MANY_DOTS 3.000
score AC_HTML_NONSENSE_TAGS 1.999
score ADMITS_SPAM 0.001
score ADVANCE_FEE_2_NEW_MONEY 1.999
-score ADVANCE_FEE_3_NEW 1.859
-score ADVANCE_FEE_3_NEW_MONEY 2.314
+score ADVANCE_FEE_3_NEW 3.108
+score ADVANCE_FEE_3_NEW_MONEY 2.030
score ADVANCE_FEE_4_NEW_MONEY 0.001
-score ADVANCE_FEE_5_NEW_FRM_MNY 2.168
-score ADVANCE_FEE_5_NEW_MONEY 0.872
-score AD_PREFS 0.001
+score ADVANCE_FEE_5_NEW_FRM_MNY 0.761
+score ADVANCE_FEE_5_NEW_MONEY 0.834
+score AD_PREFS 0.276
score AMAZON_IMG_NOT_RCVD_AMZN 0.001
score AXB_XMAILER_MIMEOLE_OL_024C2 0.001
-score BIGNUM_EMAILS_MANY 2.999
-score BITCOIN_DEADLINE 1.971
-score BITCOIN_EXTORT_01 3.491
+score BITCOIN_DEADLINE 1.957
+score BITCOIN_EXTORT_01 3.601
score BITCOIN_MALF_HTML 3.499
score BITCOIN_MALWARE 0.001
-score BITCOIN_SPAM_02 2.026
-score BITCOIN_SPAM_05 1.884
+score BITCOIN_SPAM_02 2.166
+score BITCOIN_SPAM_05 1.247
score BITCOIN_SPAM_07 0.001
-score BITCOIN_XPRIO 0.784
-score BITCOIN_YOUR_INFO 1.626
-score BODY_SINGLE_URI 0.497
-score BODY_URI_ONLY 2.765
+score BITCOIN_XPRIO 0.866
+score BITCOIN_YOUR_INFO 1.508
+score BODY_URI_ONLY 2.764
score CK_HELO_GENERIC 0.249
score CTE_8BIT_MISMATCH 0.001
-score DEAR_BENEFICIARY 2.492
+score DEAR_BENEFICIARY 2.348
score DKIMWL_BL 2.999
-score DKIMWL_WL_HIGH -0.001
-score DKIMWL_WL_MED -0.001
-score DKIMWL_WL_MEDHI -0.001 # force non-zero
+score DKIMWL_WL_HIGH -0.040
+score DKIMWL_WL_MED -0.001 # force non-zero
+score DKIMWL_WL_MEDHI -0.001
score DOS_BODY_HIGH_NO_MID 0.001
score DX_TEXT_03 1.899
score END_FUTURE_EMAILS 2.499
score FILL_THIS_FORM 0.699
score FONT_INVIS_DIRECT 0.001
-score FONT_INVIS_LONG_LINE 3.000
+score FONT_INVIS_LONG_LINE 2.999
score FONT_INVIS_MSGID 0.001
score FONT_INVIS_POSTEXTRAS 3.499
score FORGED_SPF_HELO 2.199
score FORM_FRAUD 0.999
score FORM_FRAUD_5 0.001
-score FREEMAIL_FORGED_FROMDOMAIN 0.249
-score FROMSPACE 2.199
+score FREEMAIL_FORGED_FROMDOMAIN 0.250
+score FROMSPACE 2.200
score FROM_2_EMAILS_SHORT 0.001
-score FROM_ADDR_WS 1.290
-score FROM_GOV_DKIM_AU -0.518
-score FROM_IN_TO_AND_SUBJ 1.688
-score FROM_MISSPACED 1.393
-score FROM_MISSP_EH_MATCH 1.118
-score FROM_MISSP_FREEMAIL 0.102
-score FROM_MISSP_MSFT 0.001
-score FROM_MISSP_REPLYTO 2.243
+score FROM_ADDR_WS 1.291
+score FROM_GOV_DKIM_AU -0.159
+score FROM_IN_TO_AND_SUBJ 1.819
+score FROM_MISSPACED 1.401
+score FROM_MISSP_EH_MATCH 1.213
+score FROM_MISSP_FREEMAIL 0.590
+score FROM_MISSP_MSFT 0.001 # force non-zero
+score FROM_MISSP_REPLYTO 2.493
score FROM_MISSP_SPF_FAIL 0.001
score FROM_MISSP_USER 0.001
-score FROM_MISSP_XPRIO 0.941
+score FROM_MISSP_XPRIO 0.939
score FROM_MULTI_NORDNS 0.001
-score FROM_PAYPAL_SPOOF 0.001
+score FROM_PAYPAL_SPOOF 0.081
score FROM_SUSPICIOUS_NTLD 0.001
score FROM_SUSPICIOUS_NTLD_FP 1.999
score FSL_BULK_SIG 0.001
score FSL_CTYPE_WIN1251 0.001
-score FSL_HELO_FAKE 0.001 # force non-zero
+score FSL_HELO_FAKE 0.001
score FSL_NEW_HELO_USER 0.001
score FUZZY_AMAZON 0.001
-score GB_BITCOIN_CP 2.293
-score GB_CUSTOM_HTM_URI 0.733
+score GB_BITCOIN_CP 1.955
+score GB_CUSTOM_HTM_URI 0.803
score GB_FAKE_RF_SHORT 1.999
score GB_FREEMAIL_DISPTO 0.001
score GB_HASHBL_BTC 0.001
score GOOG_REDIR_NORDNS 3.299
-score GOOG_STO_NOIMG_HTML 3.000
-score HAS_X_OUTGOING_SPAM_STAT 1.003
+score GOOG_STO_NOIMG_HTML 2.999
+score HAS_X_OUTGOING_SPAM_STAT 0.898
score HDRS_MISSP 0.001
score HDR_ORDER_FTSDMCXX_DIRECT 0.001
score HDR_ORDER_FTSDMCXX_NORDNS 0.001
score HEADER_FROM_DIFFERENT_DOMAINS 0.250
-score HELO_NO_DOMAIN 0.138
+score HELO_NO_DOMAIN 0.553
score HK_NAME_MR_MRS 0.999
score HK_RANDOM_ENVFROM 0.001
score HK_RANDOM_FROM 0.001
score HK_RANDOM_REPLYTO 0.999
-score HK_SCAM 0.746
-score HOSTED_IMG_DIRECT_MX 0.001 # force non-zero
-score HOSTED_IMG_FREEM 0.792
+score HK_SCAM 0.748
+score HOSTED_IMG_DIRECT_MX 0.001
+score HOSTED_IMG_FREEM 0.001
score HOSTED_IMG_MULTI_PUB_01 2.999
score HTML_ENTITY_ASCII 2.999
-score HTML_FONT_TINY_NORDNS 0.682
+score HTML_FONT_TINY_NORDNS 0.525
score HTML_OFF_PAGE 0.001
score HTML_SINGLET_MANY 2.499
score HTML_TAG_BALANCE_CENTER 2.799
score HTML_TEXT_INVISIBLE_FONT 1.751
score HTML_TEXT_INVISIBLE_STYLE 0.001
-score KHOP_FAKE_EBAY 0.191
score KHOP_HELO_FCRDNS 0.001 # force non-zero
score LONGLN_LOW_CONTRAST 0.001
score LONG_HEX_URI 2.999
score LONG_IMG_URI 0.001
-score LONG_INVISIBLE_TEXT 0.386
-score LOTS_OF_MONEY 0.010
+score LONG_INVISIBLE_TEXT 0.368
+score LOTS_OF_MONEY 0.009
score MALWARE_NORDNS 0.001
-score MALWARE_PASSWORD 1.082
+score MALWARE_PASSWORD 1.142
score MANY_SPAN_IN_TEXT 2.299
score MILLION_HUNDRED 0.001
-score MILLION_USD 1.357
-score MIMEOLE_DIRECT_TO_MX 0.001
-score MIXED_ES 1.999
+score MILLION_USD 1.999
+score MIMEOLE_DIRECT_TO_MX 0.001 # force non-zero
+score MIXED_ES 1.899
score MONEY_ATM_CARD 0.001
score MONEY_FORM 0.001
score MONEY_FORM_SHORT 0.001
-score MONEY_FRAUD_3 0.001 # force non-zero
-score MONEY_FRAUD_5 0.001
-score MONEY_FRAUD_8 0.001
-score MONEY_FREEMAIL_REPTO 2.337
+score MONEY_FRAUD_3 0.001
+score MONEY_FRAUD_5 0.969
+score MONEY_FRAUD_8 0.001 # force non-zero
+score MONEY_FREEMAIL_REPTO 2.387
score MONEY_FROM_MISSP 0.001
-score MSMAIL_PRI_ABNORMAL 0.229
-score NAME_EMAIL_DIFF 0.811
+score MSMAIL_PRI_ABNORMAL 1.499
+score NAME_EMAIL_DIFF 0.812
score NA_DOLLARS 1.499
-score NICE_REPLY_A -2.265
-score NO_FM_NAME_IP_HOSTN 0.291
+score NICE_REPLY_A -1.045
+score NO_FM_NAME_IP_HOSTN 0.008
score NSL_RCVD_FROM_USER 0.001
-score NSL_RCVD_HELO_USER 0.001
-score NUMBERONLY_BITCOIN_EXP 1.827
-score OBFU_TEXT_ATTACH 1.060
-score ODD_FREEM_REPTO 2.518
+score NSL_RCVD_HELO_USER 2.428
+score NUMBERONLY_BITCOIN_EXP 0.282
+score OBFU_TEXT_ATTACH 1.100
+score ODD_FREEM_REPTO 2.520
score PDS_BAD_THREAD_QP_64 0.001
-score PDS_BTC_ID 0.225
-score PDS_BTC_MSGID 0.001
+score PDS_BTC_ID 0.221
+score PDS_BTC_MSGID 0.001 # force non-zero
score PDS_FRNOM_TODOM_DBL_URL 1.499
score PDS_FRNOM_TODOM_NAKED_TO 1.499
-score PDS_FROM_2_EMAILS 2.763
+score PDS_FROM_2_EMAILS 2.720
score PDS_FROM_NAME_TO_DOMAIN 1.999
-score PDS_HP_HELO_NORDNS 0.001 # force non-zero
-score PDS_NAKED_TO_NUMERO 1.144
+score PDS_HP_HELO_NORDNS 0.001
+score PDS_NAKED_TO_NUMERO 0.179
score PDS_RDNS_DYNAMIC_FP 0.001 # force non-zero
score PDS_SHORTFWD_URISHRT_FP 1.499
score PDS_TINYSUBJ_URISHRT 1.499
score PDS_TO_EQ_FROM_NAME 0.001
-score PHP_ORIG_SCRIPT 1.924
-score PHP_SCRIPT 2.294
+score PHP_ORIG_SCRIPT 1.801
+score PHP_SCRIPT 2.399
score PP_MIME_FAKE_ASCII_TEXT 0.001
score RAND_MKTG_HEADER 1.999
score RATWARE_NO_RDNS 0.001
-score RCVD_IN_MSPIKE_H2 -0.001 # force non-zero
+score RCVD_IN_MSPIKE_H2 -0.001
score SCC_CANSPAM_2 2.699
score SCC_ISEMM_LID_1B 1.499
score SENDGRID_REDIR 1.067
score SERGIO_SUBJECT_PORN014 3.099
-score SERGIO_SUBJECT_VIAGRA01 1.223
+score SERGIO_SUBJECT_VIAGRA01 3.999
score SHOPIFY_IMG_NOT_RCVD_SFY 2.499
score SHORT_SHORTNER 1.999
-score SPOOFED_FREEMAIL 0.001 # force non-zero
-score SPOOFED_FREEMAIL_NO_RDNS 0.001
+score SPOOFED_FREEMAIL 0.001
+score SPOOFED_FREEMAIL_NO_RDNS 0.001 # force non-zero
score SPOOFED_FREEM_REPTO 0.001
score SPOOF_GMAIL_MID 0.001
-score STATIC_XPRIO_OLE 1.999
-score STOX_BOUND_090909_B 0.995
+score STATIC_XPRIO_OLE 2.000
+score STOX_BOUND_090909_B 0.883
score STY_INVIS_DIRECT 2.499
score SUSP_UTF8_WORD_COMBO 2.999
score SUSP_UTF8_WORD_FROM 1.999
-score TEQF_USR_POLITE 1.994
+score TEQF_USR_POLITE 1.993
score THIS_AD 1.999
score TONOM_EQ_TOLOC_SHRT_SHRTNER 1.499
-score TO_EQ_FM_DIRECT_MX 1.023
+score TO_EQ_FM_DIRECT_MX 1.332
score TO_EQ_FM_DOM_HTML_IMG 2.499
score TO_EQ_FM_DOM_SPF_FAIL 0.001
-score TO_EQ_FM_HTML_ONLY 2.183
+score TO_EQ_FM_HTML_ONLY 1.738
score TO_EQ_FM_SPF_FAIL 0.001
score TO_IN_SUBJ 0.099
-score TO_NAME_SUBJ_NO_RDNS 2.928
+score TO_NAME_SUBJ_NO_RDNS 2.999
score TO_NO_BRKTS_FROM_MSSP 2.499
score TO_NO_BRKTS_HTML_IMG 1.999
score TO_NO_BRKTS_HTML_ONLY 1.999
-score TO_NO_BRKTS_MSFT 1.978
+score TO_NO_BRKTS_MSFT 1.976
score TO_NO_BRKTS_NORDNS_HTML 1.999
score TO_NO_BRKTS_PCNT 2.499
score TVD_SPACE_ENCODED 0.001
score TVD_SPACE_RATIO_MINFP 0.001
score UNDISC_FREEM 2.499
-score UNDISC_MONEY 2.799
+score UNDISC_MONEY 2.800
score URI_GOOGLE_PROXY 2.199
score URI_GOOG_STO_SPAMMY 2.999
score URI_PHISH 3.999
score URI_TRY_3LD 1.999
score URI_WP_HACKED_2 2.499
-score VFY_ACCT_NORDNS 2.579
+score VFY_ACCT_NORDNS 2.563
score WORD_INVIS 2.732
score WORD_INVIS_MANY 2.999
score XM_RANDOM 0.001
@@ -212,6 +209,7 @@ score APP_DEVELOPMENT_FREEM 1.0
score APP_DEVELOPMENT_NORDNS 1.000
score BEBEE_IMG_NOT_RCVD_BB 1.000
score BIGNUM_EMAILS_FREEM 1.000
+score BIGNUM_EMAILS_MANY 1.000
score BITCOIN_BOMB 1.000
score BITCOIN_EXTORT_02 1.000
score BITCOIN_IMGUR 1.000
@@ -282,6 +280,7 @@ score GOOG_STO_HTML_PHISH_MANY 1.0
score GOOG_STO_IMG_HTML 1.000
score GOOG_STO_IMG_NOHTML 1.000
score HAS_X_NO_RELAY 1.000
+score HDRS_LCASE 0.100
score HEXHASH_WORD 1.000
score HK_CTE_RAW 1.000
score HK_RCVD_IP_MULTICAST 1.000
Modified: spamassassin/trunk/rulesrc/scores/stats-set0
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set0?rev=1911680&r1=1911679&r2=1911680&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set0 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set0 Tue Aug 15 02:42:31 2023
@@ -1,40 +1,40 @@
##### WITH NEW RULES AND SCORES #####
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 380132 54.831% (99.287% of non-spam corpus)
-# Correctly spam: 164882 23.783% (53.116% of spam corpus)
-# False positives: 2730 0.394% (0.713% of nonspam, 71929 weighted)
-# False negatives: 145535 20.992% (46.884% of spam, 272120 weighted)
-# Average score for spam: 6.4 nonspam: 0.6
+# Correctly non-spam: 378920 54.859% (99.290% of non-spam corpus)
+# Correctly spam: 164119 23.761% (53.099% of spam corpus)
+# False positives: 2710 0.392% (0.710% of nonspam, 71047 weighted)
+# False negatives: 144962 20.987% (46.901% of spam, 271468 weighted)
+# Average score for spam: 6.5 nonspam: 0.5
# Average for false-pos: 6.5 false-neg: 1.9
-# TOTAL: 693279 100.00%
+# TOTAL: 690711 100.00%
Reading scores from "tmprules"...
Reading per-message hit stat logs and scores...
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 47381 99.29%
-# Correctly spam: 20984 54.19%
-# False positives: 339 0.71%
-# False negatives: 17740 45.81%
-# TCR(l=50): 1.116287 SpamRecall: 54.189% SpamPrec: 98.410%
+# Correctly non-spam: 47229 99.29%
+# Correctly spam: 20848 54.06%
+# False positives: 336 0.71%
+# False negatives: 17719 45.94%
+# TCR(l=50): 1.117269 SpamRecall: 54.057% SpamPrec: 98.414%
##### 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: 374157 97.73%
-# Correctly spam: 94441 30.42%
-# False positives: 8705 2.27%
-# False negatives: 215976 69.58%
-# TCR(l=50): 0.476666 SpamRecall: 30.424% SpamPrec: 91.561%
+# Correctly non-spam: 373070 97.76%
+# Correctly spam: 94614 30.61%
+# False positives: 8560 2.24%
+# False negatives: 214467 69.39%
+# TCR(l=50): 0.481085 SpamRecall: 30.611% SpamPrec: 91.703%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 46641 97.74%
-# Correctly spam: 11962 30.89%
-# False positives: 1079 2.26%
-# False negatives: 26762 69.11%
-# TCR(l=50): 0.479780 SpamRecall: 30.890% SpamPrec: 91.726%
+# Correctly non-spam: 46491 97.74%
+# Correctly spam: 11770 30.52%
+# False positives: 1074 2.26%
+# False negatives: 26797 69.48%
+# TCR(l=50): 0.479111 SpamRecall: 30.518% SpamPrec: 91.638%
Modified: spamassassin/trunk/rulesrc/scores/stats-set1
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set1?rev=1911680&r1=1911679&r2=1911680&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set1 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set1 Tue Aug 15 02:42:31 2023
@@ -1,40 +1,40 @@
##### WITH NEW RULES AND SCORES #####
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 357339 82.589% (98.862% of non-spam corpus)
-# Correctly spam: 41719 9.642% (58.577% of spam corpus)
-# False positives: 4113 0.951% (1.138% of nonspam, 89714 weighted)
-# False negatives: 29502 6.819% (41.423% of spam, 32150 weighted)
-# Average score for spam: 9.9 nonspam: -0.4
-# Average for false-pos: 11.1 false-neg: 1.1
-# TOTAL: 432673 100.00%
+# Correctly non-spam: 357404 82.604% (98.881% of non-spam corpus)
+# Correctly spam: 41652 9.627% (58.483% of spam corpus)
+# False positives: 4045 0.935% (1.119% of nonspam, 90628 weighted)
+# False negatives: 29569 6.834% (41.517% of spam, 32349 weighted)
+# Average score for spam: 10.3 nonspam: -0.3
+# Average for false-pos: 11.4 false-neg: 1.1
+# TOTAL: 432670 100.00%
Reading scores from "tmprules"...
Reading per-message hit stat logs and scores...
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 44499 98.84%
+# Correctly non-spam: 44564 98.99%
# Correctly spam: 5230 58.25%
-# False positives: 520 1.16%
+# False positives: 455 1.01%
# False negatives: 3748 41.75%
-# TCR(l=50): 0.301802 SpamRecall: 58.254% SpamPrec: 90.957%
+# TCR(l=50): 0.338818 SpamRecall: 58.254% SpamPrec: 91.996%
##### 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: 358220 99.11%
+# Correctly non-spam: 358156 99.09%
# Correctly spam: 33821 47.49%
-# False positives: 3232 0.89%
+# False positives: 3293 0.91%
# False negatives: 37400 52.51%
-# TCR(l=50): 0.357894 SpamRecall: 47.487% SpamPrec: 91.277%
+# TCR(l=50): 0.352492 SpamRecall: 47.487% SpamPrec: 91.127%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 44595 99.06%
+# Correctly non-spam: 44656 99.19%
# Correctly spam: 4258 47.43%
-# False positives: 424 0.94%
+# False positives: 363 0.81%
# False negatives: 4720 52.57%
-# TCR(l=50): 0.346373 SpamRecall: 47.427% SpamPrec: 90.944%
+# TCR(l=50): 0.392567 SpamRecall: 47.427% SpamPrec: 92.145%