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 2023/07/25 02:44:41 UTC

svn commit: r1911257 [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=1911257&r1=1911256&r2=1911257&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/scores-set3 (original)
+++ spamassassin/trunk/rulesrc/scores/scores-set3 Tue Jul 25 02:44:41 2023
@@ -1,66 +1,66 @@
 # Using score set 1 logs for revision 1911193 from:
 # ham-net-ena-week0.r1911193.log ham-net-ena-week1.r1911193.log ham-net-ena-week2.r1911193.log ham-net-ena-week3.r1911193.log ham-net-ena-week4.r1911193.log ham-net-giovanni-ham.r1911193.log ham-net-giovanni-spammy.r1911193.log ham-net-giovanni-spam.r1911193.log ham-net-grenier.r1911193.log ham-net-hege.r1911193.log ham-net-jhardin.r1911193.log ham-net-llanga.r1911193.log ham-net-mmiroslaw-mails-ham.r1911193.log ham-net-mmiroslaw-mails-spam.r1911193.log ham-net-spamsponge.r1911193.log ham-net-thendrikx.r1911193.log ham-net-whyscream.r1911193.log spam-net-ena-week0.r1911193.log spam-net-ena-week1.r1911193.log spam-net-ena-week2.r1911193.log spam-net-ena-week3.r1911193.log spam-net-ena-week4.r1911193.log spam-net-giovanni-ham.r1911193.log spam-net-giovanni-spammy.r1911193.log spam-net-giovanni-spam.r1911193.log spam-net-grenier.r1911193.log spam-net-hege.r1911193.log spam-net-jhardin.r1911193.log spam-net-llanga.r1911193.log spam-net-mmiroslaw-mails-ham.r1911193.log spam-net-mmiroslaw
 -mails-spam.r1911193.log spam-net-spamsponge.r1911193.log spam-net-thendrikx.r1911193.log spam-net-whyscream.r1911193.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              2.999
 score AC_HTML_NONSENSE_TAGS          1.999
-score ADMITS_SPAM                    0.001 # force non-zero
+score ADMITS_SPAM                    0.001
 score ADVANCE_FEE_2_NEW_MONEY        1.999
-score ADVANCE_FEE_3_NEW              3.272
+score ADVANCE_FEE_3_NEW              3.267
 score ADVANCE_FEE_3_NEW_MONEY        0.001
 score ADVANCE_FEE_4_NEW_MONEY        0.001
-score ADVANCE_FEE_5_NEW_FRM_MNY      2.028
-score ADVANCE_FEE_5_NEW_MONEY        0.540
-score AD_PREFS                       0.369
+score ADVANCE_FEE_5_NEW_FRM_MNY      1.299
+score ADVANCE_FEE_5_NEW_MONEY        0.486
+score AD_PREFS                       0.107
 score AMAZON_IMG_NOT_RCVD_AMZN       0.001
-score AXB_XMAILER_MIMEOLE_OL_024C2   0.001 # force non-zero
+score AXB_XMAILER_MIMEOLE_OL_024C2   0.001
 score BIGNUM_EMAILS_MANY             2.999
-score BITCOIN_DEADLINE               1.818
-score BITCOIN_EXTORT_01              4.396
-score BITCOIN_MALF_HTML              3.311
-score BITCOIN_MALWARE                0.576
-score BITCOIN_SPAM_02                0.004
-score BITCOIN_SPAM_05                0.166
+score BITCOIN_DEADLINE               2.127
+score BITCOIN_EXTORT_01              4.443
+score BITCOIN_MALF_HTML              3.223
+score BITCOIN_MALWARE                0.606
+score BITCOIN_SPAM_02                0.001
+score BITCOIN_SPAM_05                0.756
 score BITCOIN_SPAM_07                0.001
 score BITCOIN_XPRIO                  0.001
-score BITCOIN_YOUR_INFO              2.733
+score BITCOIN_YOUR_INFO              2.409
 score BODY_SINGLE_URI                2.499
-score BODY_SINGLE_WORD               0.001 # force non-zero
-score BODY_URI_ONLY                  2.044
+score BODY_SINGLE_WORD               0.001
+score BODY_URI_ONLY                  2.046
 score CK_HELO_GENERIC                0.250
 score CTE_8BIT_MISMATCH              0.001
 score DEAR_BENEFICIARY               0.001
 score DKIMWL_BL                      2.669
 score DKIMWL_WL_HIGH                 -0.001
-score DKIMWL_WL_MED                  -0.001
+score DKIMWL_WL_MED                  -0.001 # force non-zero
 score DKIMWL_WL_MEDHI                -0.001
-score DOS_BODY_HIGH_NO_MID           0.001
+score DOS_BODY_HIGH_NO_MID           0.001 # force non-zero
 score DX_TEXT_03                     1.299
 score END_FUTURE_EMAILS              2.499
 score FILL_THIS_FORM                 0.899
-score FONT_INVIS_DIRECT              0.001 # force non-zero
+score FONT_INVIS_DIRECT              0.001
 score FONT_INVIS_LONG_LINE           2.999
 score FONT_INVIS_MSGID               0.001
-score FONT_INVIS_POSTEXTRAS          3.084
+score FONT_INVIS_POSTEXTRAS          3.083
 score FORGED_SPF_HELO                0.029
 score FORM_FRAUD                     0.999
-score FORM_FRAUD_5                   0.350
-score FREEMAIL_FORGED_FROMDOMAIN     0.249
-score FROM_2_EMAILS_SHORT            0.614
-score FROM_ADDR_WS                   1.643
+score FORM_FRAUD_5                   0.407
+score FREEMAIL_FORGED_FROMDOMAIN     0.250
+score FROM_2_EMAILS_SHORT            0.617
+score FROM_ADDR_WS                   1.639
 score FROM_FMBLA_NEWDOM28            0.799
 score FROM_GOV_DKIM_AU               -0.999
-score FROM_IN_TO_AND_SUBJ            1.691
+score FROM_IN_TO_AND_SUBJ            1.880
 score FROM_MISSPACED                 0.001
-score FROM_MISSP_EH_MATCH            0.561
-score FROM_MISSP_FREEMAIL            0.092
-score FROM_MISSP_MSFT                0.001 # force non-zero
-score FROM_MISSP_REPLYTO             0.001
+score FROM_MISSP_EH_MATCH            0.567
+score FROM_MISSP_FREEMAIL            0.095
+score FROM_MISSP_MSFT                0.001
+score FROM_MISSP_REPLYTO             0.001 # force non-zero
 score FROM_MISSP_SPF_FAIL            0.001
 score FROM_MISSP_USER                0.001
-score FROM_PAYPAL_SPOOF              1.498
+score FROM_PAYPAL_SPOOF              0.547
 score FROM_SUSPICIOUS_NTLD           0.001
 score FROM_SUSPICIOUS_NTLD_FP        1.999
 score FROM_WSP_TRAIL                 2.499
@@ -70,131 +70,131 @@ score FSL_HELO_FAKE                  0.0
 score FSL_NEW_HELO_USER              0.001
 score FUZZY_AMAZON                   0.001
 score GB_CUSTOM_HTM_URI              1.499
-score GB_FAKE_RF_SHORT               1.999
+score GB_FAKE_RF_SHORT               2.000
 score GB_FREEMAIL_DISPTO             0.001
-score GB_HASHBL_BTC                  0.001
+score GB_HASHBL_BTC                  0.216
 score GOOG_REDIR_HTML_ONLY           1.999
 score GOOG_REDIR_NORDNS              3.099
-score GOOG_STO_NOIMG_HTML            2.713
-score HAS_X_OUTGOING_SPAM_STAT       1.009
+score GOOG_STO_NOIMG_HTML            2.714
+score HAS_X_OUTGOING_SPAM_STAT       0.882
 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 HEADER_FROM_DIFFERENT_DOMAINS  0.249
 score HELO_NO_DOMAIN                 0.001
-score HK_LOTTO                       0.001
+score HK_LOTTO                       0.001 # force non-zero
 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_RANDOM_REPLYTO              1.000
 score HK_SCAM                        1.999
 score HOSTED_IMG_DIRECT_MX           0.001
-score HOSTED_IMG_FREEM               3.214
+score HOSTED_IMG_FREEM               3.216
 score HOSTED_IMG_MULTI_PUB_01        2.999
 score HTML_ENTITY_ASCII              2.999
 score HTML_FONT_TINY_NORDNS          0.001
 score HTML_OFF_PAGE                  0.001
 score HTML_SINGLET_MANY              2.499
 score HTML_TAG_BALANCE_CENTER        2.699
-score HTML_TEXT_INVISIBLE_FONT       1.748
+score HTML_TEXT_INVISIBLE_FONT       1.750
 score HTML_TEXT_INVISIBLE_STYLE      0.001
 score KHOP_HELO_FCRDNS               0.001
-score LONGLN_LOW_CONTRAST            0.001
+score LONGLN_LOW_CONTRAST            0.001 # force non-zero
 score LONG_HEX_URI                   2.999
-score LONG_IMG_URI                   0.001
+score LONG_IMG_URI                   0.001 # force non-zero
 score LONG_INVISIBLE_TEXT            0.976
 score LOTS_OF_MONEY                  0.010
 score LOTTO_DEPT                     1.999
 score MALWARE_NORDNS                 0.001
-score MALWARE_PASSWORD               1.437
+score MALWARE_PASSWORD               1.741
 score MANY_SPAN_IN_TEXT              2.099
 score MILLION_HUNDRED                0.001
 score MILLION_USD                    1.999
 score MIMEOLE_DIRECT_TO_MX           0.001
 score MIXED_ES                       1.899
-score MONEY_ATM_CARD                 0.001
+score MONEY_ATM_CARD                 0.001 # force non-zero
 score MONEY_FORM                     0.001
 score MONEY_FORM_SHORT               0.001
 score MONEY_FRAUD_3                  0.001
-score MONEY_FRAUD_5                  0.805
-score MONEY_FRAUD_8                  0.001
-score MONEY_FREEMAIL_REPTO           1.239
-score MONEY_FROM_MISSP               0.001
-score MSMAIL_PRI_ABNORMAL            0.158
+score MONEY_FRAUD_5                  0.798
+score MONEY_FRAUD_8                  0.001 # force non-zero
+score MONEY_FREEMAIL_REPTO           1.245
+score MONEY_FROM_MISSP               0.001 # force non-zero
+score MSMAIL_PRI_ABNORMAL            0.892
 score NA_DOLLARS                     1.499
 score NICE_REPLY_A                   -0.091
 score NO_FM_NAME_IP_HOSTN            0.001
 score NSL_RCVD_FROM_USER             0.001
 score NSL_RCVD_HELO_USER             0.001
-score OBFU_TEXT_ATTACH               0.346
-score ODD_FREEM_REPTO                2.552
-score PDS_BAD_THREAD_QP_64           1.000
-score PDS_BTC_ID                     0.410
+score OBFU_TEXT_ATTACH               0.340
+score ODD_FREEM_REPTO                2.551
+score PDS_BAD_THREAD_QP_64           0.999
+score PDS_BTC_ID                     0.499
 score PDS_BTC_MSGID                  0.001
 score PDS_EMPTYSUBJ_URISHRT          1.499
 score PDS_FRNOM_TODOM_DBL_URL        1.499
-score PDS_FRNOM_TODOM_NAKED_TO       1.500
-score PDS_FROM_2_EMAILS              3.000
+score PDS_FRNOM_TODOM_NAKED_TO       1.499
+score PDS_FROM_2_EMAILS              3.004
 score PDS_FROM_NAME_TO_DOMAIN        1.999
-score PDS_HP_HELO_NORDNS             0.001
+score PDS_HP_HELO_NORDNS             0.001 # force non-zero
 score PDS_OTHER_BAD_TLD              0.001
 score PDS_TINYSUBJ_URISHRT           0.001
 score PDS_TO_EQ_FROM_NAME            0.001
 score PHP_ORIG_SCRIPT                0.001
-score PHP_SCRIPT                     2.285
-score PP_MIME_FAKE_ASCII_TEXT        0.001
-score RAND_MKTG_HEADER               1.497
+score PHP_SCRIPT                     2.411
+score PP_MIME_FAKE_ASCII_TEXT        0.001 # force non-zero
+score RAND_MKTG_HEADER               1.503
 score RATWARE_NO_RDNS                0.001
 score RCVD_IN_MSPIKE_H2              -0.001
 score SCC_BODY_SINGLE_WORD           0.001
 score SCC_CANSPAM_2                  2.900
 score SCC_ISEMM_LID_1B               1.499
-score SENDGRID_REDIR                 1.038
-score SERGIO_SUBJECT_PORN014         1.374
-score SERGIO_SUBJECT_VIAGRA01        3.203
-score SHOPIFY_IMG_NOT_RCVD_SFY       1.482
+score SENDGRID_REDIR                 1.037
+score SERGIO_SUBJECT_PORN014         2.256
+score SERGIO_SUBJECT_VIAGRA01        3.337
+score SHOPIFY_IMG_NOT_RCVD_SFY       1.478
 score SHORT_SHORTNER                 1.999
 score SPOOFED_FREEMAIL               0.001
-score SPOOFED_FREEMAIL_NO_RDNS       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            1.970
-score SUBJ_BROKEN_WORD               2.399
-score SUSP_UTF8_WORD_COMBO           2.999
-score SUSP_UTF8_WORD_FROM            2.000
+score STOX_BOUND_090909_B            1.918
+score SUBJ_BROKEN_WORD               0.580
+score SUSP_UTF8_WORD_COMBO           3.000
+score SUSP_UTF8_WORD_FROM            1.999
 score THIS_AD                        1.599
 score TONOM_EQ_TOLOC_SHRT_SHRTNER    1.499
-score TO_EQ_FM_DIRECT_MX             1.222
+score TO_EQ_FM_DIRECT_MX             1.261
 score TO_EQ_FM_DOM_HTML_IMG          2.399
 score TO_EQ_FM_DOM_SPF_FAIL          0.001
 score TO_EQ_FM_SPF_FAIL              0.001 # force non-zero
 score TO_IN_SUBJ                     0.099
-score TO_NAME_SUBJ_NO_RDNS           2.956
+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          2.000
-score TO_NO_BRKTS_MSFT               0.861
-score TO_NO_BRKTS_NORDNS_HTML        1.923
+score TO_NO_BRKTS_HTML_IMG           2.000
+score TO_NO_BRKTS_HTML_ONLY          1.999
+score TO_NO_BRKTS_MSFT               0.871
+score TO_NO_BRKTS_NORDNS_HTML        1.919
 score TO_NO_BRKTS_PCNT               2.499
-score TVD_IP_HEX                     2.358
+score TVD_IP_HEX                     2.332
 score TVD_SPACE_ENCODED              0.001
 score TVD_SPACE_RATIO_MINFP          0.001
 score UNDISC_FREEM                   2.599
 score UNDISC_MONEY                   2.899
 score UNICODE_OBFU_ASC               2.499
-score URI_GOOGLE_PROXY               2.400
-score URI_ONLY_MSGID_MALF            0.001
-score URI_PHISH                      3.674
+score URI_GOOGLE_PROXY               2.399
+score URI_ONLY_MSGID_MALF            1.999
+score URI_PHISH                      3.629
 score URI_TRY_3LD                    1.999
-score URI_WP_HACKED                  0.561
+score URI_WP_HACKED                  0.562
 score URI_WP_HACKED_2                2.499
 score VFY_ACCT_NORDNS                2.999
-score WORD_INVIS                     2.342
-score WORD_INVIS_MANY                2.999
-score XFER_LOTSA_MONEY               0.988
+score WORD_INVIS                     2.344
+score WORD_INVIS_MANY                3.000
+score XFER_LOTSA_MONEY               0.985
 score XM_RANDOM                      0.001
-score XM_RECPTID                     2.999
+score XM_RECPTID                     3.000
 score XPRIO                          0.001
 score YOUR_PERMISSION                2.699
 score AC_POST_EXTRAS                 1.000

Modified: spamassassin/trunk/rulesrc/scores/stats-set0
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set0?rev=1911257&r1=1911256&r2=1911257&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set0 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set0 Tue Jul 25 02:44:41 2023
@@ -1,40 +1,40 @@
 ##### WITH NEW RULES AND SCORES #####
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam: 396883  51.219%  (99.172% of non-spam corpus)
-# Correctly spam:     183357  23.663%  (48.936% of spam corpus)
-# False positives:      3313  0.428%  (0.828% of nonspam, 110344 weighted)
-# False negatives:    191329  24.691%  (51.064% of spam, 392090 weighted)
-# Average score for spam:  6.5    nonspam: 0.7
-# Average for false-pos:   7.1  false-neg: 2.0
-# TOTAL:              774882  100.00%
+# Correctly non-spam: 396662  51.622%  (99.160% of non-spam corpus)
+# Correctly spam:     180501  23.491%  (49.000% of spam corpus)
+# False positives:      3360  0.437%  (0.840% of nonspam, 109624 weighted)
+# False negatives:    187872  24.450%  (51.000% of spam, 391073 weighted)
+# Average score for spam:  6.6    nonspam: 0.7
+# Average for false-pos:   7.1  false-neg: 2.1
+# TOTAL:              768395  100.00%
 
 Reading scores from "tmprules"...
 Reading per-message hit stat logs and scores...
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam:  49538  99.21%
-# Correctly spam:      22609  48.43%
-# False positives:       395  0.79%
-# False negatives:     24078  51.57%
-# TCR(l=50): 1.065232  SpamRecall: 48.427%  SpamPrec: 98.283%
+# Correctly non-spam:  49511  99.20%
+# Correctly spam:      22370  48.73%
+# False positives:       400  0.80%
+# False negatives:     23535  51.27%
+# TCR(l=50): 1.054439  SpamRecall: 48.731%  SpamPrec: 98.243%
 
 ##### 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: 391224  97.76%
-# Correctly spam:     118598  31.65%
-# False positives:      8972  2.24%
-# False negatives:    256088  68.35%
-# TCR(l=50): 0.531705  SpamRecall: 31.653%  SpamPrec: 92.967%
+# Correctly non-spam: 390972  97.74%
+# Correctly spam:     118201  32.09%
+# False positives:      9050  2.26%
+# False negatives:    250172  67.91%
+# TCR(l=50): 0.524246  SpamRecall: 32.087%  SpamPrec: 92.888%
 Reading scores from "../rules-base"...
 Reading per-message hit stat logs and scores...
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam:  48826  97.78%
-# Correctly spam:      14852  31.81%
-# False positives:      1107  2.22%
-# False negatives:     31835  68.19%
-# TCR(l=50): 0.535493  SpamRecall: 31.812%  SpamPrec: 93.063%
+# Correctly non-spam:  48787  97.75%
+# Correctly spam:      14677  31.97%
+# False positives:      1124  2.25%
+# False negatives:     31228  68.03%
+# TCR(l=50): 0.525061  SpamRecall: 31.973%  SpamPrec: 92.887%

Modified: spamassassin/trunk/rulesrc/scores/stats-set1
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set1?rev=1911257&r1=1911256&r2=1911257&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set1 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set1 Tue Jul 25 02:44:41 2023
@@ -1,11 +1,11 @@
 ##### WITH NEW RULES AND SCORES #####
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam: 373871  76.547%  (98.914% of non-spam corpus)
-# Correctly spam:      78742  16.122%  (71.297% of spam corpus)
-# False positives:      4105  0.840%  (1.086% of nonspam, 136612 weighted)
-# False negatives:     31701  6.491%  (28.703% of spam,  27075 weighted)
-# Average score for spam:  13.8    nonspam: -0.4
+# Correctly non-spam: 373870  76.547%  (98.914% of non-spam corpus)
+# Correctly spam:      78774  16.128%  (71.325% of spam corpus)
+# False positives:      4106  0.841%  (1.086% of nonspam, 136635 weighted)
+# False negatives:     31669  6.484%  (28.675% of spam,  26920 weighted)
+# Average score for spam:  14.1    nonspam: -0.4
 # Average for false-pos:   11.4  false-neg: 0.9
 # TOTAL:              488419  100.00%
 
@@ -13,11 +13,11 @@ Reading scores from "tmprules"...
 Reading per-message hit stat logs and scores...
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam:  46601  98.90%
+# Correctly non-spam:  46599  98.89%
 # Correctly spam:       9891  70.85%
-# False positives:       519  1.10%
+# False positives:       521  1.11%
 # False negatives:      4070  29.15%
-# TCR(l=50): 0.465057  SpamRecall: 70.847%  SpamPrec: 95.014%
+# TCR(l=50): 0.463513  SpamRecall: 70.847%  SpamPrec: 94.996%
 
 ##### WITHOUT NEW RULES AND SCORES #####
 Reading scores from "../rules-base"...