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 2023/07/28 02:45:42 UTC

svn commit: r1911319 [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=1911319&r1=1911318&r2=1911319&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/scores-set3 (original)
+++ spamassassin/trunk/rulesrc/scores/scores-set3 Fri Jul 28 02:45:41 2023
@@ -1,82 +1,81 @@
 # 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             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
 score ADVANCE_FEE_2_NEW_MONEY        1.999
-score ADVANCE_FEE_3_NEW              3.266
-score ADVANCE_FEE_3_NEW_MONEY        0.001 # force non-zero
+score ADVANCE_FEE_3_NEW              3.268
+score ADVANCE_FEE_3_NEW_MONEY        0.001
 score ADVANCE_FEE_4_NEW_MONEY        0.001
-score ADVANCE_FEE_5_NEW_FRM_MNY      1.383
-score ADVANCE_FEE_5_NEW_MONEY        0.590
+score ADVANCE_FEE_5_NEW_FRM_MNY      1.570
+score ADVANCE_FEE_5_NEW_MONEY        0.394
 score AD_PREFS                       0.001
-score AMAZON_IMG_NOT_RCVD_AMZN       0.001
+score AMAZON_IMG_NOT_RCVD_AMZN       0.001 # force non-zero
 score AXB_XMAILER_MIMEOLE_OL_024C2   0.001
 score BIGNUM_EMAILS_MANY             2.999
-score BITCOIN_DEADLINE               2.446
-score BITCOIN_EXTORT_01              4.193
-score BITCOIN_MALF_HTML              3.406
-score BITCOIN_MALWARE                0.058
-score BITCOIN_SPAM_02                0.001
-score BITCOIN_SPAM_05                1.678
-score BITCOIN_SPAM_07                0.001 # force non-zero
-score BITCOIN_XPRIO                  0.001
-score BITCOIN_YOUR_INFO              2.999
+score BITCOIN_DEADLINE               2.052
+score BITCOIN_EXTORT_01              4.496
+score BITCOIN_MALF_HTML              3.226
+score BITCOIN_MALWARE                0.452
+score BITCOIN_SPAM_02                0.003
+score BITCOIN_SPAM_05                0.516
+score BITCOIN_SPAM_07                0.001
+score BITCOIN_XPRIO                  0.001 # force non-zero
+score BITCOIN_YOUR_INFO              2.779
 score BODY_SINGLE_URI                2.499
-score BODY_SINGLE_WORD               0.001 # force non-zero
-score BODY_URI_ONLY                  2.047
+score BODY_SINGLE_WORD               0.001
+score BODY_URI_ONLY                  2.048
 score CK_HELO_GENERIC                0.250
 score CTE_8BIT_MISMATCH              0.001
 score DEAR_BENEFICIARY               0.001 # force non-zero
-score DKIMWL_BL                      2.672
+score DKIMWL_BL                      2.670
 score DKIMWL_WL_HIGH                 -0.001
 score DKIMWL_WL_MED                  -0.001
-score DKIMWL_WL_MEDHI                -0.001 # force non-zero
+score DKIMWL_WL_MEDHI                -0.001
 score DOS_BODY_HIGH_NO_MID           0.001
 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          3.085
-score FORGED_SPF_HELO                0.030
+score FONT_INVIS_MSGID               0.001 # force non-zero
+score FONT_INVIS_POSTEXTRAS          3.078
+score FORGED_SPF_HELO                0.029
 score FORM_FRAUD                     0.999
-score FORM_FRAUD_5                   0.299
+score FORM_FRAUD_5                   0.495
 score FREEMAIL_FORGED_FROMDOMAIN     0.250
-score FROM_2_EMAILS_SHORT            0.617
-score FROM_ADDR_WS                   1.642
+score FROM_2_EMAILS_SHORT            0.620
+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            1.722
-score FROM_MISSPACED                 0.001
-score FROM_MISSP_EH_MATCH            0.559
-score FROM_MISSP_FREEMAIL            0.096
+score FROM_IN_TO_AND_SUBJ            1.737
+score FROM_MISSPACED                 1.020
+score FROM_MISSP_EH_MATCH            0.556
+score FROM_MISSP_FREEMAIL            0.097
 score FROM_MISSP_MSFT                0.001
 score FROM_MISSP_REPLYTO             0.001
 score FROM_MISSP_SPF_FAIL            0.001
 score FROM_MISSP_USER                0.001
-score FROM_PAYPAL_SPOOF              0.094
+score FROM_PAYPAL_SPOOF              1.133
 score FROM_SUSPICIOUS_NTLD           0.001
 score FROM_SUSPICIOUS_NTLD_FP        1.999
 score FROM_WSP_TRAIL                 2.499
 score FSL_BULK_SIG                   0.001
 score FSL_CTYPE_WIN1251              0.001
-score FSL_HELO_FAKE                  0.001
+score FSL_HELO_FAKE                  0.001 # force non-zero
 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_FREEMAIL_DISPTO             0.001
-score GB_HASHBL_BTC                  0.349
-score GOOG_REDIR_HTML_ONLY           1.999
+score GB_HASHBL_BTC                  0.001
+score GOOG_REDIR_HTML_ONLY           2.000
 score GOOG_REDIR_NORDNS              3.099
-score GOOG_STO_NOIMG_HTML            2.685
-score HAS_X_OUTGOING_SPAM_STAT       0.897
+score GOOG_STO_NOIMG_HTML            2.714
+score HAS_X_OUTGOING_SPAM_STAT       0.899
 score HDRS_MISSP                     0.001
 score HDR_ORDER_FTSDMCXX_DIRECT      0.001
 score HDR_ORDER_FTSDMCXX_NORDNS      0.001
@@ -85,116 +84,116 @@ score HELO_NO_DOMAIN                 0.0
 score HK_LOTTO                       0.001
 score HK_NAME_MR_MRS                 0.999
 score HK_RANDOM_ENVFROM              0.001
-score HK_RANDOM_FROM                 0.001 # force non-zero
-score HK_RANDOM_REPLYTO              0.999
+score HK_RANDOM_FROM                 0.001
+score HK_RANDOM_REPLYTO              1.000
 score HK_SCAM                        1.999
-score HOSTED_IMG_DIRECT_MX           0.001
-score HOSTED_IMG_FREEM               3.213
+score HOSTED_IMG_DIRECT_MX           0.001 # force non-zero
+score HOSTED_IMG_FREEM               3.214
 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       1.752
+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 LONG_HEX_URI                   2.999
 score LONG_IMG_URI                   0.001
-score LONG_INVISIBLE_TEXT            0.972
+score LONG_INVISIBLE_TEXT            0.975
 score LOTS_OF_MONEY                  0.010
 score LOTTO_DEPT                     1.999
 score MALWARE_NORDNS                 0.001
-score MALWARE_PASSWORD               1.694
+score MALWARE_PASSWORD               1.514
 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 MIMEOLE_DIRECT_TO_MX           0.001 # force non-zero
+score MIXED_ES                       1.900
 score MONEY_ATM_CARD                 0.001
-score MONEY_FORM                     0.001
+score MONEY_FORM                     0.001 # force non-zero
 score MONEY_FORM_SHORT               0.001
 score MONEY_FRAUD_3                  0.001
-score MONEY_FRAUD_5                  0.801
+score MONEY_FRAUD_5                  0.799
 score MONEY_FRAUD_8                  0.001
 score MONEY_FREEMAIL_REPTO           1.238
 score MONEY_FROM_MISSP               0.001
-score MSMAIL_PRI_ABNORMAL            0.247
+score MSMAIL_PRI_ABNORMAL            0.332
 score NA_DOLLARS                     1.499
-score NICE_REPLY_A                   -0.091
+score NICE_REPLY_A                   -0.094
 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               0.348
-score ODD_FREEM_REPTO                2.544
-score PDS_BTC_ID                     0.316
+score NSL_RCVD_HELO_USER             0.001 # force non-zero
+score OBFU_TEXT_ATTACH               0.344
+score ODD_FREEM_REPTO                2.548
+score PDS_BTC_ID                     0.496
 score PDS_BTC_MSGID                  0.001
-score PDS_EMPTYSUBJ_URISHRT          1.499
+score PDS_EMPTYSUBJ_URISHRT          1.500
 score PDS_FRNOM_TODOM_DBL_URL        1.499
 score PDS_FRNOM_TODOM_NAKED_TO       1.499
-score PDS_FROM_2_EMAILS              3.000
-score PDS_FROM_NAME_TO_DOMAIN        2.000
+score PDS_FROM_2_EMAILS              3.001
+score PDS_FROM_NAME_TO_DOMAIN        1.999
 score PDS_HP_HELO_NORDNS             0.001
-score PDS_OTHER_BAD_TLD              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 # force non-zero
-score PHP_SCRIPT                     2.398
-score PP_MIME_FAKE_ASCII_TEXT        0.001
-score RAND_MKTG_HEADER               1.508
+score PHP_ORIG_SCRIPT                0.001
+score PHP_SCRIPT                     2.393
+score PP_MIME_FAKE_ASCII_TEXT        0.001 # force non-zero
+score RAND_MKTG_HEADER               1.500
 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_BODY_SINGLE_WORD           0.001 # force non-zero
+score SCC_CANSPAM_2                  2.899
 score SCC_ISEMM_LID_1B               1.499
-score SENDGRID_REDIR                 1.037
-score SERGIO_SUBJECT_PORN014         1.843
-score SERGIO_SUBJECT_VIAGRA01        1.349
-score SHOPIFY_IMG_NOT_RCVD_SFY       1.485
+score SENDGRID_REDIR                 1.038
+score SERGIO_SUBJECT_PORN014         1.596
+score SERGIO_SUBJECT_VIAGRA01        2.617
+score SHOPIFY_IMG_NOT_RCVD_SFY       1.484
 score SHORT_SHORTNER                 1.999
 score SPOOFED_FREEMAIL               0.001
 score SPOOFED_FREEMAIL_NO_RDNS       0.001
 score SPOOFED_FREEM_REPTO            0.001
 score SPOOF_GMAIL_MID                0.001
 score STATIC_XPRIO_OLE               1.999
-score SUBJ_BROKEN_WORD               1.087
+score SUBJ_BROKEN_WORD               0.231
 score SUSP_UTF8_WORD_COMBO           2.999
 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             1.261
+score TO_EQ_FM_DIRECT_MX             1.227
 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           2.911
+score TO_EQ_FM_SPF_FAIL              0.001
+score TO_IN_SUBJ                     0.100
+score TO_NAME_SUBJ_NO_RDNS           2.941
 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               0.870
-score TO_NO_BRKTS_NORDNS_HTML        1.925
+score TO_NO_BRKTS_HTML_ONLY          1.999
+score TO_NO_BRKTS_MSFT               0.871
+score TO_NO_BRKTS_NORDNS_HTML        1.922
 score TO_NO_BRKTS_PCNT               2.499
-score TVD_IP_HEX                     2.380
+score TVD_IP_HEX                     2.446
 score TVD_SPACE_ENCODED              0.001
 score TVD_SPACE_RATIO_MINFP          0.001
 score UNDISC_FREEM                   2.599
-score UNDISC_MONEY                   2.900
+score UNDISC_MONEY                   2.899
 score UNICODE_OBFU_ASC               2.499
 score URI_GOOGLE_PROXY               2.399
-score URI_ONLY_MSGID_MALF            1.196
-score URI_PHISH                      3.634
+score URI_ONLY_MSGID_MALF            1.185
+score URI_PHISH                      3.669
 score URI_TRY_3LD                    1.999
-score URI_WP_HACKED                  0.566
-score URI_WP_HACKED_2                2.499
+score URI_WP_HACKED                  0.562
+score URI_WP_HACKED_2                2.500
 score VFY_ACCT_NORDNS                2.999
-score WORD_INVIS                     2.343
+score WORD_INVIS                     2.347
 score WORD_INVIS_MANY                2.999
-score XFER_LOTSA_MONEY               0.984
+score XFER_LOTSA_MONEY               0.987
 score XM_RANDOM                      0.001
 score XM_RECPTID                     2.999
 score XPRIO                          0.001
-score YOUR_PERMISSION                2.615
+score YOUR_PERMISSION                2.476
 score AC_POST_EXTRAS                 1.000
 score AC_SPAMMY_URI_PATTERNS1        1.000
 score AC_SPAMMY_URI_PATTERNS10       1.000

Modified: spamassassin/trunk/rulesrc/scores/stats-set0
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set0?rev=1911319&r1=1911318&r2=1911319&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set0 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set0 Fri Jul 28 02:45:41 2023
@@ -1,40 +1,40 @@
 ##### WITH NEW RULES AND SCORES #####
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam: 396807  52.940%  (99.155% of non-spam corpus)
-# Correctly spam:     171495  22.880%  (49.089% of spam corpus)
-# False positives:      3382  0.451%  (0.845% of nonspam, 104892 weighted)
-# False negatives:    177857  23.729%  (50.911% of spam, 374325 weighted)
-# Average score for spam:  6.7    nonspam: 0.7
-# Average for false-pos:   7.1  false-neg: 2.1
-# TOTAL:              749541  100.00%
+# Correctly non-spam: 394860  53.737%  (99.121% of non-spam corpus)
+# Correctly spam:     170584  23.215%  (50.704% of spam corpus)
+# False positives:      3503  0.477%  (0.879% of nonspam,  99822 weighted)
+# False negatives:    165850  22.571%  (49.296% of spam, 346093 weighted)
+# Average score for spam:  6.5    nonspam: 0.7
+# Average for false-pos:   6.7  false-neg: 2.1
+# TOTAL:              734797  100.00%
 
 Reading scores from "tmprules"...
 Reading per-message hit stat logs and scores...
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam:  49512  99.16%
-# Correctly spam:      21459  49.28%
-# False positives:       421  0.84%
-# False negatives:     22085  50.72%
-# TCR(l=50): 1.009482  SpamRecall: 49.281%  SpamPrec: 98.076%
+# Correctly non-spam:  49286  99.19%
+# Correctly spam:      21390  51.00%
+# False positives:       403  0.81%
+# False negatives:     20553  49.00%
+# TCR(l=50): 1.030465  SpamRecall: 50.998%  SpamPrec: 98.151%
 
 ##### 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: 391215  97.76%
-# Correctly spam:     114942  32.90%
-# False positives:      8974  2.24%
-# False negatives:    234410  67.10%
-# TCR(l=50): 0.511414  SpamRecall: 32.901%  SpamPrec: 92.758%
+# Correctly non-spam: 389249  97.71%
+# Correctly spam:     112814  33.53%
+# False positives:      9114  2.29%
+# False negatives:    223620  66.47%
+# TCR(l=50): 0.495251  SpamRecall: 33.532%  SpamPrec: 92.525%
 Reading scores from "../rules-base"...
 Reading per-message hit stat logs and scores...
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam:  48815  97.76%
-# Correctly spam:      14483  33.26%
-# False positives:      1118  2.24%
-# False negatives:     29061  66.74%
-# TCR(l=50): 0.512518  SpamRecall: 33.261%  SpamPrec: 92.834%
+# Correctly non-spam:  48608  97.82%
+# Correctly spam:      14256  33.99%
+# False positives:      1081  2.18%
+# False negatives:     27687  66.01%
+# TCR(l=50): 0.513146  SpamRecall: 33.989%  SpamPrec: 92.952%

Modified: spamassassin/trunk/rulesrc/scores/stats-set1
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set1?rev=1911319&r1=1911318&r2=1911319&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set1 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set1 Fri Jul 28 02:45:41 2023
@@ -1,11 +1,11 @@
 ##### WITH NEW RULES AND SCORES #####
 
 # SUMMARY for threshold 5.0:
-# Correctly non-spam: 373875  76.548%  (98.915% of non-spam corpus)
-# Correctly spam:      78767  16.127%  (71.319% of spam corpus)
-# False positives:      4101  0.840%  (1.085% of nonspam, 136557 weighted)
-# False negatives:     31676  6.485%  (28.681% of spam,  26953 weighted)
-# Average score for spam:  13.7    nonspam: -0.4
+# Correctly non-spam: 373867  76.546%  (98.913% of non-spam corpus)
+# Correctly spam:      78723  16.118%  (71.279% of spam corpus)
+# False positives:      4109  0.841%  (1.087% of nonspam, 136696 weighted)
+# False negatives:     31720  6.494%  (28.721% of spam,  27127 weighted)
+# Average score for spam:  13.8    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:  46603  98.90%
-# Correctly spam:       9892  70.85%
-# False positives:       517  1.10%
-# False negatives:      4069  29.15%
-# TCR(l=50): 0.466627  SpamRecall: 70.855%  SpamPrec: 95.033%
+# Correctly non-spam:  46599  98.89%
+# Correctly spam:       9887  70.82%
+# False positives:       521  1.11%
+# False negatives:      4074  29.18%
+# TCR(l=50): 0.463451  SpamRecall: 70.819%  SpamPrec: 94.994%
 
 ##### WITHOUT NEW RULES AND SCORES #####
 Reading scores from "../rules-base"...