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 2023/04/14 02:40:22 UTC
svn commit: r1909129 [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=1909129&r1=1909128&r2=1909129&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/scores-set3 (original)
+++ spamassassin/trunk/rulesrc/scores/scores-set3 Fri Apr 14 02:40:21 2023
@@ -1,192 +1,191 @@
# Using score set 1 logs for revision 1909022 from:
# ham-net-ena-week0.r1909022.log ham-net-ena-week1.r1909022.log ham-net-ena-week2.r1909022.log ham-net-giovanni-ham.r1909022.log ham-net-giovanni-spammy.r1909022.log ham-net-giovanni-spam.r1909022.log ham-net-grenier.r1909022.log ham-net-hege.r1909022.log ham-net-jhardin.r1909022.log ham-net-llanga.r1909022.log ham-net-mmiroslaw-mails-ham.r1909022.log ham-net-mmiroslaw-mails-spam.r1909022.log ham-net-spamsponge.r1909022.log ham-net-thendrikx.r1909022.log spam-net-ena-week0.r1909022.log spam-net-ena-week1.r1909022.log spam-net-ena-week2.r1909022.log spam-net-giovanni-ham.r1909022.log spam-net-giovanni-spammy.r1909022.log spam-net-giovanni-spam.r1909022.log spam-net-grenier.r1909022.log spam-net-hege.r1909022.log spam-net-jhardin.r1909022.log spam-net-llanga.r1909022.log spam-net-mmiroslaw-mails-ham.r1909022.log spam-net-mmiroslaw-mails-spam.r1909022.log spam-net-spamsponge.r1909022.log spam-net-thendrikx.r1909022.log
-score ACCT_PHISHING_MANY 3.000
+score ACCT_PHISHING_MANY 2.999
score AC_DIV_BONANZA 0.001
score AC_FROM_MANY_DOTS 2.999
score AC_HTML_NONSENSE_TAGS 1.999
score ADMITS_SPAM 0.001
-score ADVANCE_FEE_2_NEW_MONEY 2.000
-score ADVANCE_FEE_3_NEW 1.627
+score ADVANCE_FEE_2_NEW_MONEY 1.999
+score ADVANCE_FEE_3_NEW 1.635
score ADVANCE_FEE_3_NEW_MONEY 0.001
-score ADVANCE_FEE_4_NEW_FRM_MNY 1.117
-score ADVANCE_FEE_4_NEW_MONEY 1.857
-score ADVANCE_FEE_5_NEW_FRM_MNY 0.975
+score ADVANCE_FEE_4_NEW_FRM_MNY 0.820
+score ADVANCE_FEE_4_NEW_MONEY 1.730
+score ADVANCE_FEE_5_NEW_FRM_MNY 0.927
score ADVANCE_FEE_5_NEW_MONEY 0.001
-score AD_PREFS 0.302
-score AMAZON_IMG_NOT_RCVD_AMZN 0.001 # force non-zero
+score AD_PREFS 0.289
+score AMAZON_IMG_NOT_RCVD_AMZN 0.001
score AXB_XMAILER_MIMEOLE_OL_024C2 0.001
-score BITCOIN_DEADLINE 1.955
-score BITCOIN_EXTORT_01 2.414
-score BITCOIN_MALF_HTML 1.333
-score BITCOIN_MALWARE 1.276
-score BITCOIN_OBFU_SUBJ 1.769
-score BITCOIN_SPAM_02 2.077
-score BITCOIN_SPAM_05 1.156
-score BITCOIN_SPAM_07 2.339
+score BITCOIN_DEADLINE 2.999
+score BITCOIN_EXTORT_01 2.078
+score BITCOIN_MALF_HTML 0.996
+score BITCOIN_MALWARE 0.403
+score BITCOIN_OBFU_SUBJ 1.766
+score BITCOIN_SPAM_02 1.915
+score BITCOIN_SPAM_05 1.558
+score BITCOIN_SPAM_07 2.669
score BITCOIN_XPRIO 0.001
score BITCOIN_YOUR_INFO 2.999
-score BODY_SINGLE_URI 1.950
-score BODY_URI_ONLY 0.534
+score BODY_SINGLE_URI 1.947
+score BODY_URI_ONLY 0.547
score CK_HELO_GENERIC 0.250
score CTE_8BIT_MISMATCH 0.001
-score DEAR_BENEFICIARY 1.968
-score DKIMWL_BL 2.666
+score DEAR_BENEFICIARY 2.351
+score DKIMWL_BL 2.662
score DKIMWL_WL_HIGH -0.001
score DKIMWL_WL_MED -0.001
-score DKIMWL_WL_MEDHI -0.001
-score END_FUTURE_EMAILS 2.500
-score FILL_THIS_FORM 0.800
-score FONT_INVIS_DIRECT 0.001
-score FONT_INVIS_MSGID 0.001
-score FORGED_SPF_HELO 0.029
+score DKIMWL_WL_MEDHI -0.230
+score END_FUTURE_EMAILS 2.499
+score FILL_THIS_FORM 0.799
+score FONT_INVIS_DIRECT 0.001 # force non-zero
+score FONT_INVIS_MSGID 0.001 # force non-zero
+score FORGED_SPF_HELO 0.030
score FORM_FRAUD 0.747
score FORM_FRAUD_5 0.001
score FREEMAIL_FORGED_FROMDOMAIN 0.250
score FROM_FMBLA_NEWDOM 0.001
score FROM_FMBLA_NEWDOM14 0.001
score FROM_FMBLA_NEWDOM28 0.799
-score FROM_GOV_DKIM_AU -0.816
+score FROM_GOV_DKIM_AU -0.248
score FROM_IN_TO_AND_SUBJ 1.899
score FROM_MISSPACED 1.999
-score FROM_MISSP_EH_MATCH 0.398
+score FROM_MISSP_EH_MATCH 0.399
score FROM_MISSP_FREEMAIL 0.001
score FROM_MISSP_MSFT 0.001
-score FROM_MISSP_REPLYTO 1.610
+score FROM_MISSP_REPLYTO 1.600
score FROM_MISSP_SPF_FAIL 0.001
score FROM_MISSP_USER 0.001 # force non-zero
-score FROM_NTLD_LINKBAIT 0.764
-score FROM_PAYPAL_SPOOF 1.012
-score FROM_SUSPICIOUS_NTLD 0.001
+score FROM_NTLD_LINKBAIT 0.490
+score FROM_PAYPAL_SPOOF 1.334
+score FROM_SUSPICIOUS_NTLD 0.001 # force non-zero
score FROM_SUSPICIOUS_NTLD_FP 1.999
score FROM_WSP_TRAIL 2.399
-score FSL_BULK_SIG 0.001 # force non-zero
+score FSL_BULK_SIG 0.001
score FSL_CTYPE_WIN1251 0.001
score FSL_NEW_HELO_USER 0.001
score FUZZY_AMAZON 0.001
score FUZZY_PAYPAL 1.899
-score FUZZY_WALLET 0.001
-score GB_BITCOIN_CP 2.435
+score FUZZY_WALLET 0.001 # force non-zero
+score GB_BITCOIN_CP 1.012
score GB_CUSTOM_HTM_URI 0.836
-score GB_FAKE_RF_SHORT 0.829
+score GB_FAKE_RF_SHORT 0.831
score GB_FREEMAIL_DISPTO 0.001
score GB_GOOGLE_TRANSL 0.750
-score GB_HASHBL_BTC 1.194
-score GOOG_REDIR_HTML_ONLY 1.999
+score GB_HASHBL_BTC 1.322
+score GOOG_REDIR_HTML_ONLY 2.000
score GOOG_REDIR_NORDNS 3.099
-score GOOG_STO_NOIMG_HTML 1.378
+score GOOG_STO_NOIMG_HTML 1.376
score HAS_X_OUTGOING_SPAM_STAT 0.001
score HDRS_LCASE_IMGONLY 0.100
-score HDRS_MISSP 0.741
+score HDRS_MISSP 0.740
score HDR_ORDER_FTSDMCXX_DIRECT 0.001
score HDR_ORDER_FTSDMCXX_NORDNS 0.001
-score HEADER_FROM_DIFFERENT_DOMAINS 0.249
+score HEADER_FROM_DIFFERENT_DOMAINS 0.250
score HELO_NO_DOMAIN 0.001
score HK_LOTTO 0.999
score HK_NAME_MR_MRS 1.000
-score HK_RANDOM_ENVFROM 0.001
-score HK_RANDOM_FROM 0.059
+score HK_RANDOM_ENVFROM 0.001 # force non-zero
+score HK_RANDOM_FROM 0.058
score HK_RANDOM_REPLYTO 0.999
score HK_SCAM 1.999
-score HOSTED_IMG_DIRECT_MX 0.001 # force non-zero
-score HOSTED_IMG_FREEM 1.593
-score HTML_ENTITY_ASCII_TINY 2.999
-score HTML_OFF_PAGE 0.001
-score HTML_SINGLET_MANY 2.415
-score HTML_TEXT_INVISIBLE_FONT 2.000
+score HOSTED_IMG_DIRECT_MX 0.001
+score HOSTED_IMG_FREEM 1.589
+score HTML_ENTITY_ASCII_TINY 3.000
+score HTML_OFF_PAGE 0.001 # force non-zero
+score HTML_SINGLET_MANY 2.417
+score HTML_TEXT_INVISIBLE_FONT 1.999
score HTML_TEXT_INVISIBLE_STYLE 1.722
score JH_SPAMMY_HEADERS 3.499
-score KHOP_HELO_FCRDNS 0.001
+score KHOP_HELO_FCRDNS 0.001 # force non-zero
score LONG_HEX_URI 1.615
score LONG_IMG_URI 0.001
-score LONG_INVISIBLE_TEXT 2.997
+score LONG_INVISIBLE_TEXT 2.999
score LOTS_OF_MONEY 0.010
score MALWARE_NORDNS 0.001
-score MALWARE_PASSWORD 0.425
+score MALWARE_PASSWORD 0.001
score MANY_SPAN_IN_TEXT 3.299
-score MILLION_HUNDRED 1.081
+score MILLION_HUNDRED 1.083
score MILLION_USD 1.999
score MIMEOLE_DIRECT_TO_MX 0.001
score MIXED_ES 1.999
-score MONEY_ATM_CARD 2.023
-score MONEY_FORM 2.235
-score MONEY_FORM_SHORT 0.089
-score MONEY_FRAUD_3 2.499
+score MONEY_ATM_CARD 1.814
+score MONEY_FORM 2.721
+score MONEY_FORM_SHORT 0.090
+score MONEY_FRAUD_3 2.500
score MONEY_FRAUD_5 0.001
score MONEY_FRAUD_8 0.001
-score MONEY_FREEMAIL_REPTO 1.500
+score MONEY_FREEMAIL_REPTO 1.492
score MONEY_FROM_MISSP 0.001
-score MSMAIL_PRI_ABNORMAL 0.502
-score NICE_REPLY_A -1.220
+score MSMAIL_PRI_ABNORMAL 0.514
+score NICE_REPLY_A -2.282
score NO_FM_NAME_IP_HOSTN 0.001
score NSL_RCVD_FROM_USER 0.001
score NSL_RCVD_HELO_USER 0.001
-score NUMBERONLY_BITCOIN_EXP 0.605
+score NUMBERONLY_BITCOIN_EXP 0.001
score OBFU_TEXT_ATTACH 1.447
-score ODD_FREEM_REPTO 2.552
-score PDS_BAD_THREAD_QP_64 1.000
-score PDS_BTC_ID 0.114
-score PDS_BTC_MSGID 0.001
-score PDS_BTC_NTLD 0.956
+score ODD_FREEM_REPTO 2.744
+score PDS_BAD_THREAD_QP_64 0.999
+score PDS_BTC_ID 0.324
+score PDS_BTC_MSGID 0.001 # force non-zero
+score PDS_BTC_NTLD 1.584
score PDS_EMPTYSUBJ_URISHRT 1.499
-score PDS_FRNOM_TODOM_DBL_URL 0.552
+score PDS_FRNOM_TODOM_DBL_URL 0.716
score PDS_FRNOM_TODOM_NAKED_TO 1.499
score PDS_FROM_2_EMAILS_SHRTNER 1.499
score PDS_FROM_NAME_TO_DOMAIN 1.999
-score PDS_HELO_SPF_FAIL 0.067
-score PDS_NAKED_TO_NUMERO 0.755
-score PDS_NO_FULL_NAME_SPOOFED_URL 0.750
+score PDS_HELO_SPF_FAIL 0.118
+score PDS_NAKED_TO_NUMERO 1.287
+score PDS_NO_FULL_NAME_SPOOFED_URL 0.749
score PDS_SHORT_SPOOFED_URL 1.999
-score PHP_SCRIPT 2.500
-score POSSIBLE_AMAZON_PHISH_02 2.500
+score PHP_SCRIPT 2.499
+score POSSIBLE_AMAZON_PHISH_02 2.499
score POSSIBLE_GMAIL_PHISHER 2.899
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
-score SCC_CANSPAM_2 0.680
-score SCC_ISEMM_LID_1B 1.499
-score SENDGRID_REDIR 1.068
-score SERGIO_SUBJECT_VIAGRA01 0.799
-score SHOPIFY_IMG_NOT_RCVD_SFY 1.696
+score RCVD_IN_MSPIKE_H2 -0.001 # force non-zero
+score SCC_CANSPAM_2 2.419
+score SCC_ISEMM_LID_1B 1.500
+score SENDGRID_REDIR 1.066
+score SERGIO_SUBJECT_VIAGRA01 1.949
+score SHOPIFY_IMG_NOT_RCVD_SFY 1.689
score SPOOFED_FREEMAIL 0.001
score SPOOFED_FREEMAIL_NO_RDNS 0.001
-score SPOOFED_FREEM_REPTO 0.001
+score SPOOFED_FREEM_REPTO 0.001 # force non-zero
score SPOOF_GMAIL_MID 0.001
score STATIC_XPRIO_OLE 1.999
-score STOX_BOUND_090909_B 1.346
-score SUSP_UTF8_WORD_COMBO 2.415
+score STOX_BOUND_090909_B 1.191
score SUSP_UTF8_WORD_SUBJ 1.999
score THIS_AD 1.099
-score TO_EQ_FM_DIRECT_MX 0.253
+score TO_EQ_FM_DIRECT_MX 0.254
score TO_EQ_FM_DOM_SPF_FAIL 0.001
score TO_EQ_FM_SPF_FAIL 0.001
-score TO_IN_SUBJ 0.099
-score TO_NAME_SUBJ_NO_RDNS 2.999
-score TO_NO_BRKTS_FROM_MSSP 2.500
+score TO_IN_SUBJ 0.100
+score TO_NAME_SUBJ_NO_RDNS 3.000
+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 0.001
-score TO_NO_BRKTS_NORDNS_HTML 1.920
+score TO_NO_BRKTS_NORDNS_HTML 1.923
score TO_NO_BRKTS_PCNT 2.499
score TVD_SPACE_ENCODED 0.001
score UNDISC_FREEM 2.999
score UNDISC_MONEY 3.099
score URI_DOTEDU 1.999
-score URI_GIBB_NO_PROTO 2.517
+score URI_GIBB_NO_PROTO 0.540
score URI_GOOGLE_PROXY 2.299
score URI_ONLY_MSGID_MALF 0.001
-score URI_PHISH 3.825
-score URI_TRY_3LD 1.999
+score URI_PHISH 3.820
+score URI_TRY_3LD 2.000
score URI_WP_HACKED 3.499
score URI_WP_HACKED_2 2.499
score VFY_ACCT_NORDNS 2.999
-score WORD_INVIS 1.943
-score WORD_INVIS_MANY 2.994
-score XFER_LOTSA_MONEY 0.001
+score WORD_INVIS 1.942
+score WORD_INVIS_MANY 2.996
+score XFER_LOTSA_MONEY 0.001 # force non-zero
score XM_RANDOM 0.001 # force non-zero
score XPRIO 0.001
-score YOU_INHERIT 2.399
+score YOU_INHERIT 1.211
score AC_BR_BONANZA 0.001
score AC_POST_EXTRAS 1.000
score AC_SPAMMY_URI_PATTERNS1 1.000
@@ -230,6 +229,7 @@ score BTC_ORG 1.0
score BULK_RE_SUSP_NTLD 1.000
score CANT_SEE_AD 1.000
score COMMENT_GIBBERISH 1.000
+score COMPENSATION 1.000
score CONTENT_AFTER_HTML 1.000
score CONTENT_AFTER_HTML_WEAK 1.000
score DAY_I_EARNED 1.000
@@ -292,6 +292,7 @@ score LIST_PRTL_PUMPDUMP 1.0
score LIST_PRTL_SAME_USER 1.000
score LUCRATIVE 1.000
score MALF_HTML_B64 1.000
+score MANY_HDRS_LCASE 0.100
score MIME_NO_TEXT 1.000
score MIXED_AREA_CASE 1.000
score MIXED_CENTER_CASE 1.000
Modified: spamassassin/trunk/rulesrc/scores/stats-set0
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set0?rev=1909129&r1=1909128&r2=1909129&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set0 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set0 Fri Apr 14 02:40:21 2023
@@ -1,40 +1,40 @@
##### WITH NEW RULES AND SCORES #####
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 288692 67.424% (99.395% of non-spam corpus)
-# Correctly spam: 74252 17.342% (53.914% of spam corpus)
-# False positives: 1757 0.410% (0.605% of nonspam, 28618 weighted)
-# False negatives: 63471 14.824% (46.086% of spam, 136781 weighted)
-# Average score for spam: 7.9 nonspam: 0.6
-# Average for false-pos: 6.9 false-neg: 2.2
-# TOTAL: 428172 100.00%
+# Correctly non-spam: 290154 67.466% (99.382% of non-spam corpus)
+# Correctly spam: 78315 18.210% (56.702% of spam corpus)
+# False positives: 1805 0.420% (0.618% of nonspam, 26808 weighted)
+# False negatives: 59802 13.905% (43.298% of spam, 126376 weighted)
+# Average score for spam: 8.0 nonspam: 0.6
+# Average for false-pos: 6.3 false-neg: 2.1
+# TOTAL: 430076 100.00%
Reading scores from "tmprules"...
Reading per-message hit stat logs and scores...
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 36086 99.43%
-# Correctly spam: 9382 54.37%
-# False positives: 207 0.57%
-# False negatives: 7874 45.63%
-# TCR(l=50): 0.946883 SpamRecall: 54.369% SpamPrec: 97.841%
+# Correctly non-spam: 36228 99.33%
+# Correctly spam: 9619 55.59%
+# False positives: 243 0.67%
+# False negatives: 7683 44.41%
+# TCR(l=50): 0.872384 SpamRecall: 55.595% SpamPrec: 97.536%
##### 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: 283770 97.70%
-# Correctly spam: 50426 36.61%
-# False positives: 6679 2.30%
-# False negatives: 87297 63.39%
-# TCR(l=50): 0.326941 SpamRecall: 36.614% SpamPrec: 88.304%
+# Correctly non-spam: 285246 97.70%
+# Correctly spam: 50276 36.40%
+# False positives: 6713 2.30%
+# False negatives: 87841 63.60%
+# TCR(l=50): 0.326139 SpamRecall: 36.401% SpamPrec: 88.221%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 35485 97.77%
-# Correctly spam: 6303 36.53%
-# False positives: 808 2.23%
-# False negatives: 10953 63.47%
-# TCR(l=50): 0.336027 SpamRecall: 36.526% SpamPrec: 88.637%
+# Correctly non-spam: 35596 97.60%
+# Correctly spam: 6328 36.57%
+# False positives: 875 2.40%
+# False negatives: 10974 63.43%
+# TCR(l=50): 0.316168 SpamRecall: 36.574% SpamPrec: 87.852%
Modified: spamassassin/trunk/rulesrc/scores/stats-set1
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set1?rev=1909129&r1=1909128&r2=1909129&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set1 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set1 Fri Apr 14 02:40:21 2023
@@ -1,23 +1,23 @@
##### WITH NEW RULES AND SCORES #####
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 259342 79.143% (99.007% of non-spam corpus)
-# Correctly spam: 40202 12.268% (61.147% of spam corpus)
-# False positives: 2600 0.793% (0.993% of nonspam, 72781 weighted)
-# False negatives: 25545 7.796% (38.853% of spam, 41648 weighted)
-# Average score for spam: 13.3 nonspam: -0.3
-# Average for false-pos: 11.2 false-neg: 1.6
+# Correctly non-spam: 259326 79.138% (99.001% of non-spam corpus)
+# Correctly spam: 40178 12.261% (61.110% of spam corpus)
+# False positives: 2616 0.798% (0.999% of nonspam, 72987 weighted)
+# False negatives: 25569 7.803% (38.890% of spam, 41772 weighted)
+# Average score for spam: 13.4 nonspam: -0.3
+# Average for false-pos: 11.1 false-neg: 1.6
# TOTAL: 327689 100.00%
Reading scores from "tmprules"...
Reading per-message hit stat logs and scores...
# SUMMARY for threshold 5.0:
-# Correctly non-spam: 32412 98.99%
-# Correctly spam: 5065 61.21%
-# False positives: 331 1.01%
-# False negatives: 3210 38.79%
-# TCR(l=50): 0.418775 SpamRecall: 61.208% SpamPrec: 93.866%
+# Correctly non-spam: 32413 98.99%
+# Correctly spam: 5064 61.20%
+# False positives: 330 1.01%
+# False negatives: 3211 38.80%
+# TCR(l=50): 0.419816 SpamRecall: 61.196% SpamPrec: 93.882%
##### WITHOUT NEW RULES AND SCORES #####
Reading scores from "../rules-base"...