Mailing List Archive

svn commit: r1914667 [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=1914667&r1=1914666&r2=1914667&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/scores-set3 (original)
+++ spamassassin/trunk/rulesrc/scores/scores-set3 Fri Dec 15 02:45:23 2023
@@ -1,94 +1,94 @@
# Using score set 1 logs for revision 1914480 from:
# ham-net-ena-week0.r1914480.log ham-net-ena-week1.r1914480.log ham-net-ena-week2.r1914480.log ham-net-ena-week3.r1914480.log ham-net-ena-week4.r1914480.log ham-net-giovanni-ham.r1914480.log ham-net-giovanni-spammy.r1914480.log ham-net-giovanni-spam.r1914480.log ham-net-grenier.r1914480.log ham-net-hege.r1914480.log ham-net-jhardin.r1914480.log ham-net-llanga.r1914480.log ham-net-mmiroslaw-mails-ham.r1914480.log ham-net-mmiroslaw-mails-spam.r1914480.log ham-net-spamsponge.r1914480.log ham-net-whyscream.r1914480.log spam-net-ena-week0.r1914480.log spam-net-ena-week1.r1914480.log spam-net-ena-week2.r1914480.log spam-net-ena-week3.r1914480.log spam-net-ena-week4.r1914480.log spam-net-giovanni-ham.r1914480.log spam-net-giovanni-spammy.r1914480.log spam-net-giovanni-spam.r1914480.log spam-net-grenier.r1914480.log spam-net-hege.r1914480.log spam-net-jhardin.r1914480.log spam-net-llanga.r1914480.log spam-net-mmiroslaw-mails-ham.r1914480.log spam-net-mmiroslaw-mails-spam.r1914480.log spam-n
et-spamsponge.r1914480.log spam-net-whyscream.r1914480.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 2.499
+score AC_FROM_MANY_DOTS 2.500
score AC_HTML_NONSENSE_TAGS 1.999
-score AC_POST_EXTRAS 0.318
-score ADMITS_SPAM 2.299
+score AC_POST_EXTRAS 0.312
+score ADMITS_SPAM 2.399
score ADVANCE_FEE_2_NEW_MONEY 1.999
score ADVANCE_FEE_3_NEW 3.499
score ADVANCE_FEE_3_NEW_MONEY 0.001
-score ADVANCE_FEE_4_NEW 2.199
-score ADVANCE_FEE_4_NEW_MONEY 0.377
+score ADVANCE_FEE_4_NEW 2.299
+score ADVANCE_FEE_4_NEW_MONEY 0.382
score ADVANCE_FEE_5_NEW_FRM_MNY 0.001
-score ADVANCE_FEE_5_NEW_MONEY 1.332
-score AD_PREFS 0.066
+score ADVANCE_FEE_5_NEW_MONEY 1.252
+score AD_PREFS 0.129
score AXB_XMAILER_MIMEOLE_OL_024C2 0.001
-score BITCOIN_EXTORT_01 0.725
+score BITCOIN_EXTORT_01 0.665
score BITCOIN_MALF_HTML 0.001
-score BITCOIN_ONAN 2.454
-score BITCOIN_PAY_ME 1.149
-score BITCOIN_SPAM_02 1.222
-score BITCOIN_SPAM_05 0.001
-score BITCOIN_SPAM_07 0.369
-score BITCOIN_TOEQFM 3.163
+score BITCOIN_ONAN 2.782
+score BITCOIN_PAY_ME 1.532
+score BITCOIN_SPAM_02 1.061
+score BITCOIN_SPAM_05 0.001 # force non-zero
+score BITCOIN_SPAM_07 1.889
+score BITCOIN_TOEQFM 1.697
score BITCOIN_VISTA 0.001
-score BITCOIN_XPRIO 1.124
+score BITCOIN_XPRIO 1.055
score BITCOIN_YOUR_INFO 2.999
-score BODY_SINGLE_URI 0.001 # force non-zero
+score BODY_SINGLE_URI 0.001
score BODY_SINGLE_WORD 0.001
-score BODY_URI_ONLY 2.268
-score CK_HELO_GENERIC 0.249
-score CTE_8BIT_MISMATCH 0.001
-score DEAR_BENEFICIARY 0.602
+score BODY_URI_ONLY 2.270
+score CK_HELO_GENERIC 0.250
+score CTE_8BIT_MISMATCH 0.001 # force non-zero
+score DEAR_BENEFICIARY 0.691
score DKIMWL_WL_HIGH -0.001
score DKIMWL_WL_MED -0.001 # force non-zero
score DKIMWL_WL_MEDHI -0.001
-score DYNAMIC_IMGUR 3.249
-score END_FUTURE_EMAILS 2.499
-score FILL_THIS_FORM 0.600
+score DYNAMIC_IMGUR 3.547
+score END_FUTURE_EMAILS 2.500
+score FILL_THIS_FORM 0.599
score FONT_INVIS_DIRECT 0.001
-score FONT_INVIS_LONG_LINE 3.000
-score FONT_INVIS_MSGID 1.905
-score FONT_INVIS_NORDNS 1.292
+score FONT_INVIS_LONG_LINE 2.999
+score FONT_INVIS_MSGID 1.944
+score FONT_INVIS_NORDNS 1.289
score FONT_INVIS_POSTEXTRAS 3.499
-score FORGED_SPF_HELO 2.799
-score FORM_FRAUD_5 2.647
+score FORGED_SPF_HELO 2.899
+score FORM_FRAUD_5 2.699
score FREEMAIL_FORGED_FROMDOMAIN 0.249
score FROM_ADDR_WS 2.999
-score FROM_FMBLA_NEWDOM28 0.799
-score FROM_GOV_DKIM_AU -0.392
-score FROM_IN_TO_AND_SUBJ 1.191
+score FROM_FMBLA_NEWDOM28 0.800
+score FROM_GOV_DKIM_AU -0.325
+score FROM_IN_TO_AND_SUBJ 1.190
score FROM_MISSPACED 1.999
score FROM_MISSP_DYNIP 0.001
score FROM_MISSP_EH_MATCH 1.300
-score FROM_MISSP_FREEMAIL 1.952
+score FROM_MISSP_FREEMAIL 1.951
score FROM_MISSP_MSFT 1.349
-score FROM_MISSP_REPLYTO 1.882
-score FROM_MISSP_SPF_FAIL 1.034
-score FROM_MISSP_TO_UNDISC 3.324
+score FROM_MISSP_REPLYTO 1.891
+score FROM_MISSP_SPF_FAIL 1.999
+score FROM_MISSP_TO_UNDISC 3.599
score FROM_MISSP_USER 0.001
score FROM_MISSP_XPRIO 2.499
-score FROM_PAYPAL_SPOOF 0.531
-score FROM_SUSPICIOUS_NTLD 0.500
-score FROM_SUSPICIOUS_NTLD_FP 1.999
-score FROM_UNBAL1 2.200
+score FROM_PAYPAL_SPOOF 1.522
+score FROM_SUSPICIOUS_NTLD 0.499
+score FROM_SUSPICIOUS_NTLD_FP 2.000
+score FROM_UNBAL1 2.299
score FROM_WSP_TRAIL 2.399
-score FSL_BULK_SIG 2.074
-score FSL_CTYPE_WIN1251 0.428
-score FSL_HAS_TINYURL 2.299
+score FSL_BULK_SIG 2.055
+score FSL_CTYPE_WIN1251 0.259
+score FSL_HAS_TINYURL 2.300
score FSL_NEW_HELO_USER 0.001
-score FUZZY_AMAZON 2.499
-score FUZZY_BITCOIN 1.180
+score FUZZY_AMAZON 2.399
+score FUZZY_BITCOIN 1.593
score FUZZY_WALLET 1.899
score GB_BITCOIN_CP 0.001
score GB_CUSTOM_HTM_URI 1.499
-score GB_FAKE_RF_SHORT 1.127
-score GB_FREEMAIL_DISPTO 0.001
-score GB_HASHBL_BTC 1.038
+score GB_FAKE_RF_SHORT 1.123
+score GB_FREEMAIL_DISPTO 0.001 # force non-zero
+score GB_HASHBL_BTC 1.265
score GOOG_REDIR_NORDNS 0.001
score GOOG_STO_IMG_HTML 2.999
score GOOG_STO_NOIMG_HTML 2.999
score HAS_X_OUTGOING_SPAM_STAT 0.001
-score HDRS_LCASE_IMGONLY 0.099
+score HDRS_LCASE_IMGONLY 0.100
score HDRS_MISSP 2.499
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 0.250
+score HELO_MISC_IP 0.249
score HELO_NO_DOMAIN 0.001
score HK_NAME_FM_MR_MRS 1.499
score HK_NAME_MR_MRS 0.999
@@ -96,112 +96,113 @@ score HK_RANDOM_ENVFROM 0.9
score HK_RANDOM_FROM 0.999
score HK_RANDOM_REPLYTO 0.999
score HK_SCAM 1.999
-score HOSTED_IMG_DIRECT_MX 3.499
-score HOSTED_IMG_FREEM 3.016
+score HOSTED_IMG_DIRECT_MX 3.500
+score HOSTED_IMG_FREEM 3.012
score HOSTED_IMG_MULTI_PUB_01 2.999
-score HTML_BADATTR 1.000
+score HTML_BADATTR 0.999
score HTML_ENTITY_ASCII_TINY 2.999
score HTML_FONT_TINY_NORDNS 0.001
-score HTML_SINGLET_MANY 2.499
-score HTML_TEXT_INVISIBLE_FONT 1.747
-score HTML_TEXT_INVISIBLE_STYLE 2.533
+score HTML_SINGLET_MANY 2.500
+score HTML_TEXT_INVISIBLE_FONT 1.706
+score HTML_TEXT_INVISIBLE_STYLE 2.336
score JH_SPAMMY_HEADERS 3.499
score KHOP_HELO_FCRDNS 0.260
score LONG_HEX_URI 2.999
-score LONG_IMG_URI 2.941
+score LONG_IMG_URI 2.943
score LONG_INVISIBLE_TEXT 2.699
score LOTS_OF_MONEY 0.010
score LOTTO_DEPT 1.999
score MALFORMED_FREEMAIL 2.699
-score MALWARE_NORDNS 0.985
-score MANY_SUBDOM 2.200
-score MILLION_HUNDRED 0.397
-score MILLION_USD 1.999
+score MALWARE_NORDNS 0.358
+score MANY_SUBDOM 2.199
+score MILLION_HUNDRED 0.478
+score MILLION_USD 2.000
score MIMEOLE_DIRECT_TO_MX 0.001
-score MIME_NO_TEXT 0.602
+score MIME_NO_TEXT 0.034
score MIXED_ES 2.099
score MIXED_HREF_CASE 1.999
-score MONEY_ATM_CARD 1.000
+score MONEY_ATM_CARD 1.276
score MONEY_FORM 0.001
score MONEY_FORM_SHORT 1.893
score MONEY_FRAUD_3 0.001
-score MONEY_FRAUD_5 1.449
-score MONEY_FRAUD_8 2.233
-score MONEY_FREEMAIL_REPTO 1.316
+score MONEY_FRAUD_5 1.533
+score MONEY_FRAUD_8 2.014
+score MONEY_FREEMAIL_REPTO 1.313
score MONEY_FROM_MISSP 0.001
-score MONEY_NOHTML 1.263
+score MONEY_NOHTML 1.177
score NA_DOLLARS 1.499
-score NICE_REPLY_A -2.012
-score NORDNS_LOW_CONTRAST 0.001 # force non-zero
-score NO_FM_NAME_IP_HOSTN 0.001
+score NICE_REPLY_A -3.799
+score NORDNS_LOW_CONTRAST 0.001
+score NO_FM_NAME_IP_HOSTN 0.001 # force non-zero
score NSL_RCVD_FROM_USER 0.001
-score NSL_RCVD_HELO_USER 0.001
-score OBFU_BITCOIN 0.778
+score NSL_RCVD_HELO_USER 0.182
+score OBFU_BITCOIN 0.569
score OBFU_TEXT_ATTACH 1.599
-score ODD_FREEM_REPTO 2.597
+score ODD_FREEM_REPTO 2.592
score PDS_BAD_THREAD_QP_64 0.999
-score PDS_BTC_ID 0.499
-score PDS_BTC_MSGID 0.001 # force non-zero
+score PDS_BTC_ID 0.485
+score PDS_BTC_MSGID 0.001
score PDS_EMPTYSUBJ_URISHRT 1.499
score PDS_FRNOM_TODOM_DBL_URL 0.001
score PDS_FRNOM_TODOM_NAKED_TO 1.499
+score PDS_FROM_2_EMAILS 2.307
score PDS_FROM_NAME_TO_DOMAIN 1.999
score PDS_HELO_SPF_FAIL 1.999
score PDS_HP_HELO_NORDNS 0.001
score PDS_NAKED_TO_NUMERO 1.999
score PDS_OTHER_BAD_TLD 1.999
score PDS_TINYSUBJ_URISHRT 1.499
-score PHP_SCRIPT 2.500
+score PHP_SCRIPT 2.499
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_BODY_SINGLE_WORD 0.001
-score SCC_CANSPAM_2 1.188
+score SCC_CANSPAM_2 0.841
score SCC_ISEMM_LID_1B 1.499
-score SENDGRID_REDIR 0.600
+score SENDGRID_REDIR 0.617
score SHOPIFY_IMG_NOT_RCVD_SFY 2.499
-score SHORTENED_URL_SRC 2.399
+score SHORTENED_URL_SRC 2.400
score SHORT_SHORTNER 1.999
score SPOOFED_FREEMAIL 0.001
-score SPOOFED_FREEMAIL_NO_RDNS 0.001 # force non-zero
-score SPOOFED_FREEM_REPTO 1.893
-score STATIC_XPRIO_OLE 0.768
+score SPOOFED_FREEMAIL_NO_RDNS 0.001
+score SPOOFED_FREEM_REPTO 1.105
+score STATIC_XPRIO_OLE 0.766
score SUBJ_ATTENTION 0.499
-score THIS_AD 0.700
-score TONOM_EQ_TOLOC_SHRT_SHRTNER 1.499
-score TO_EQ_FM_DIRECT_MX 2.499
-score TO_EQ_FM_DOM_SPF_FAIL 0.478
-score TO_EQ_FM_SPF_FAIL 0.899
+score THIS_AD 0.799
+score TONOM_EQ_TOLOC_SHRT_SHRTNER 1.500
+score TO_EQ_FM_DIRECT_MX 2.500
+score TO_EQ_FM_DOM_SPF_FAIL 0.301
+score TO_EQ_FM_SPF_FAIL 1.066
score TO_IN_SUBJ 0.099
-score TO_NO_BRKTS_FROM_MSSP 2.499
+score TO_NO_BRKTS_FROM_MSSP 2.500
score TO_NO_BRKTS_HTML_IMG 1.999
score TO_NO_BRKTS_HTML_ONLY 1.999
-score TO_NO_BRKTS_MSFT 2.479
+score TO_NO_BRKTS_MSFT 2.478
score TO_NO_BRKTS_NORDNS_HTML 1.999
-score TO_NO_BRKTS_PCNT 2.329
-score TVD_SPACE_RATIO_MINFP 0.001 # force non-zero
+score TO_NO_BRKTS_PCNT 2.220
+score TVD_SPACE_RATIO_MINFP 0.001
score UNDISC_FREEM 2.599
score UNDISC_MONEY 2.899
score UNICODE_OBFU_ZW_MANY 2.999
-score URI_DOTEDU 1.040
+score URI_DOTEDU 0.527
score URI_GOOGLE_PROXY 1.699
-score URI_IN_URI_5 3.199
+score URI_IN_URI_5 3.299
score URI_PHISH 3.999
-score URI_PHP_REDIR 2.488
+score URI_PHP_REDIR 2.487
score URI_TRY_3LD 1.999
-score URI_WP_HACKED_2 2.500
+score URI_WP_HACKED_2 2.499
score VFY_ACCT_NORDNS 2.999
score VISTA_COST 2.499
score VISTA_TONOM_EQ_TOLOC 2.499
score WIKI_IMG 1.999
-score WORD_INVIS 1.897
+score WORD_INVIS 1.898
score WORD_INVIS_MANY 2.999
score XFER_LOTSA_MONEY 0.001
-score XM_DIGITS_ONLY 2.229
+score XM_DIGITS_ONLY 2.052
score XM_RANDOM 2.499
score XM_UC_ONLY 2.499
-score XPRIO 0.084
+score XPRIO 0.001
score AC_SPAMMY_URI_PATTERNS1 1.000
score AC_SPAMMY_URI_PATTERNS10 1.000
score AC_SPAMMY_URI_PATTERNS11 1.000

Modified: spamassassin/trunk/rulesrc/scores/stats-set0
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set0?rev=1914667&r1=1914666&r2=1914667&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set0 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set0 Fri Dec 15 02:45:23 2023
@@ -1,40 +1,40 @@
##### WITH NEW RULES AND SCORES #####

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 455062 62.774% (99.188% of non-spam corpus)
-# Correctly spam: 163975 22.620% (61.615% of spam corpus)
-# False positives: 3726 0.514% (0.812% of nonspam, 71550 weighted)
-# False negatives: 102153 14.092% (38.385% of spam, 210470 weighted)
-# Average score for spam: 6.5 nonspam: 0.6
-# Average for false-pos: 6.6 false-neg: 2.1
-# TOTAL: 724916 100.00%
+# Correctly non-spam: 452577 62.547% (99.195% of non-spam corpus)
+# Correctly spam: 165673 22.896% (61.973% of spam corpus)
+# False positives: 3673 0.508% (0.805% of nonspam, 71466 weighted)
+# False negatives: 101656 14.049% (38.027% of spam, 203665 weighted)
+# Average score for spam: 6.5 nonspam: 0.7
+# Average for false-pos: 6.6 false-neg: 2.0
+# TOTAL: 723579 100.00%

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

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 56727 99.15%
-# Correctly spam: 20350 61.13%
-# False positives: 488 0.85%
-# False negatives: 12939 38.87%
-# TCR(l=50): 0.891534 SpamRecall: 61.131% SpamPrec: 97.658%
+# Correctly non-spam: 56426 99.21%
+# Correctly spam: 20548 61.46%
+# False positives: 450 0.79%
+# False negatives: 12883 38.54%
+# TCR(l=50): 0.944832 SpamRecall: 61.464% SpamPrec: 97.857%

##### 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: 446241 97.27%
-# Correctly spam: 108793 40.88%
-# False positives: 12547 2.73%
-# False negatives: 157335 59.12%
-# TCR(l=50): 0.339153 SpamRecall: 40.880% SpamPrec: 89.660%
+# Correctly non-spam: 443756 97.26%
+# Correctly spam: 109198 40.85%
+# False positives: 12494 2.74%
+# False negatives: 158131 59.15%
+# TCR(l=50): 0.341490 SpamRecall: 40.848% SpamPrec: 89.733%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 55533 97.06%
-# Correctly spam: 13537 40.67%
-# False positives: 1682 2.94%
-# False negatives: 19752 59.33%
-# TCR(l=50): 0.320543 SpamRecall: 40.665% SpamPrec: 88.948%
+# Correctly non-spam: 55350 97.32%
+# Correctly spam: 13536 40.49%
+# False positives: 1526 2.68%
+# False negatives: 19895 59.51%
+# TCR(l=50): 0.347534 SpamRecall: 40.489% SpamPrec: 89.869%

Modified: spamassassin/trunk/rulesrc/scores/stats-set1
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set1?rev=1914667&r1=1914666&r2=1914667&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set1 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set1 Fri Dec 15 02:45:23 2023
@@ -1,23 +1,23 @@
##### WITH NEW RULES AND SCORES #####

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 430764 85.339% (99.395% of non-spam corpus)
-# Correctly spam: 26806 5.311% (37.552% of spam corpus)
-# False positives: 2623 0.520% (0.605% of nonspam, 30355 weighted)
-# False negatives: 44577 8.831% (62.448% of spam, 67988 weighted)
-# Average score for spam: 5.5 nonspam: -0.9
+# Correctly non-spam: 430725 85.331% (99.386% of non-spam corpus)
+# Correctly spam: 26907 5.331% (37.694% of spam corpus)
+# False positives: 2662 0.527% (0.614% of nonspam, 30695 weighted)
+# False negatives: 44475 8.811% (62.306% of spam, 67654 weighted)
+# Average score for spam: 5.5 nonspam: -1.0
# Average for false-pos: 7.0 false-neg: 1.5
-# TOTAL: 504770 100.00%
+# TOTAL: 504769 100.00%

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

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 53681 99.39%
-# Correctly spam: 3406 37.85%
-# False positives: 328 0.61%
-# False negatives: 5592 62.15%
-# TCR(l=50): 0.409149 SpamRecall: 37.853% SpamPrec: 91.216%
+# Correctly non-spam: 53676 99.38%
+# Correctly spam: 3396 37.74%
+# False positives: 333 0.62%
+# False negatives: 5602 62.26%
+# TCR(l=50): 0.404368 SpamRecall: 37.742% SpamPrec: 91.070%

##### 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: 431752 99.62%
-# Correctly spam: 13717 19.22%
+# Correctly spam: 13749 19.26%
# False positives: 1635 0.38%
-# False negatives: 57666 80.78%
-# TCR(l=50): 0.512014 SpamRecall: 19.216% SpamPrec: 89.350%
+# False negatives: 57633 80.74%
+# TCR(l=50): 0.512128 SpamRecall: 19.261% SpamPrec: 89.372%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
# Correctly non-spam: 53807 99.63%
-# Correctly spam: 1750 19.45%
+# Correctly spam: 1737 19.30%
# False positives: 202 0.37%
-# False negatives: 7248 80.55%
-# TCR(l=50): 0.518677 SpamRecall: 19.449% SpamPrec: 89.652%
+# False negatives: 7261 80.70%
+# TCR(l=50): 0.518288 SpamRecall: 19.304% SpamPrec: 89.582%