Mailing List Archive

svn commit: r1914377 [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=1914377&r1=1914376&r2=1914377&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/scores-set3 (original)
+++ spamassassin/trunk/rulesrc/scores/scores-set3 Wed Dec 6 02:44:44 2023
@@ -1,85 +1,85 @@
# Using score set 1 logs for revision 1914280 from:
# ham-net-ena-week0.r1914280.log ham-net-ena-week1.r1914280.log ham-net-ena-week2.r1914280.log ham-net-ena-week3.r1914280.log ham-net-ena-week4.r1914280.log ham-net-giovanni-ham.r1914280.log ham-net-giovanni-spammy.r1914280.log ham-net-giovanni-spam.r1914280.log ham-net-grenier.r1914280.log ham-net-hege.r1914280.log ham-net-jhardin.r1914280.log ham-net-llanga.r1914280.log ham-net-mmiroslaw-mails-ham.r1914280.log ham-net-mmiroslaw-mails-spam.r1914280.log ham-net-spamsponge.r1914280.log ham-net-whyscream.r1914280.log spam-net-ena-week0.r1914280.log spam-net-ena-week1.r1914280.log spam-net-ena-week2.r1914280.log spam-net-ena-week3.r1914280.log spam-net-ena-week4.r1914280.log spam-net-giovanni-ham.r1914280.log spam-net-giovanni-spammy.r1914280.log spam-net-giovanni-spam.r1914280.log spam-net-grenier.r1914280.log spam-net-hege.r1914280.log spam-net-jhardin.r1914280.log spam-net-llanga.r1914280.log spam-net-mmiroslaw-mails-ham.r1914280.log spam-net-mmiroslaw-mails-spam.r1914280.log spam-n
et-spamsponge.r1914280.log spam-net-whyscream.r1914280.log

-score ACCT_PHISHING_MANY 2.999
+score ACCT_PHISHING_MANY 3.000
score AC_BR_BONANZA 0.001
score AC_DIV_BONANZA 0.001
score AC_FROM_MANY_DOTS 2.499
score AC_HTML_NONSENSE_TAGS 1.999
-score AC_POST_EXTRAS 2.499
+score AC_POST_EXTRAS 2.500
score ADMITS_SPAM 2.399
-score ADVANCE_FEE_2_NEW_MONEY 1.992
-score ADVANCE_FEE_3_NEW 3.264
+score ADVANCE_FEE_2_NEW_MONEY 1.999
+score ADVANCE_FEE_3_NEW 3.265
score ADVANCE_FEE_3_NEW_MONEY 0.001
-score ADVANCE_FEE_4_NEW_MONEY 0.131
+score ADVANCE_FEE_4_NEW_MONEY 0.135
score ADVANCE_FEE_5_NEW_FRM_MNY 0.001
-score ADVANCE_FEE_5_NEW_MONEY 2.565
-score AD_PREFS 0.084
+score ADVANCE_FEE_5_NEW_MONEY 2.542
+score AD_PREFS 0.215
score AXB_XMAILER_MIMEOLE_OL_024C2 0.001
-score BITCOIN_DEADLINE 2.742
-score BITCOIN_EXTORT_01 0.001
-score BITCOIN_MALF_HTML 2.647
+score BITCOIN_DEADLINE 2.716
+score BITCOIN_EXTORT_01 0.112
+score BITCOIN_MALF_HTML 2.754
score BITCOIN_ONAN 2.999
-score BITCOIN_PAY_ME 0.449
-score BITCOIN_SPAM_02 1.329
+score BITCOIN_PAY_ME 1.069
+score BITCOIN_SPAM_02 1.451
score BITCOIN_SPAM_05 0.001
-score BITCOIN_SPAM_07 0.825
-score BITCOIN_TOEQFM 3.093
+score BITCOIN_SPAM_07 2.236
+score BITCOIN_TOEQFM 1.750
score BITCOIN_VISTA 0.001
-score BITCOIN_XPRIO 1.416
-score BITCOIN_YOUR_INFO 2.119
+score BITCOIN_XPRIO 1.462
+score BITCOIN_YOUR_INFO 2.127
score BODY_SINGLE_WORD 0.001
-score BODY_URI_ONLY 2.270
+score BODY_URI_ONLY 2.274
score CK_HELO_GENERIC 0.249
score CTE_8BIT_MISMATCH 0.001
-score DEAR_BENEFICIARY 0.401
+score DEAR_BENEFICIARY 0.398
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 END_FUTURE_EMAILS 2.499
score FILL_THIS_FORM 0.900
score FONT_INVIS_DIRECT 0.001
-score FONT_INVIS_LONG_LINE 2.999
-score FONT_INVIS_MSGID 1.255
-score FONT_INVIS_POSTEXTRAS 1.310
+score FONT_INVIS_LONG_LINE 3.000
+score FONT_INVIS_MSGID 1.239
+score FONT_INVIS_POSTEXTRAS 1.305
score FORGED_SPF_HELO 2.599
score FORM_FRAUD_5 2.533
score FREEMAIL_FORGED_FROMDOMAIN 0.250
score FROM_ADDR_WS 2.999
score FROM_FMBLA_NEWDOM28 0.001
-score FROM_GOV_DKIM_AU -0.005
-score FROM_IN_TO_AND_SUBJ 0.655
-score FROM_MISSPACED 1.711
+score FROM_GOV_DKIM_AU -0.999
+score FROM_IN_TO_AND_SUBJ 1.343
+score FROM_MISSPACED 1.701
score FROM_MISSP_EH_MATCH 1.999
score FROM_MISSP_FREEMAIL 2.499
-score FROM_MISSP_MSFT 0.526
-score FROM_MISSP_REPLYTO 1.601
-score FROM_MISSP_SPF_FAIL 1.999
+score FROM_MISSP_MSFT 0.518
+score FROM_MISSP_REPLYTO 1.604
+score FROM_MISSP_SPF_FAIL 1.916
score FROM_MISSP_USER 0.001
score FROM_MISSP_XPRIO 2.499
-score FROM_PAYPAL_SPOOF 1.599
+score FROM_PAYPAL_SPOOF 0.720
score FROM_SUSPICIOUS_NTLD 0.500
score FROM_SUSPICIOUS_NTLD_FP 1.999
score FROM_WSP_TRAIL 2.499
-score FSL_BULK_SIG 1.044
+score FSL_BULK_SIG 1.041
score FSL_CTYPE_WIN1251 0.001
score FSL_HAS_TINYURL 2.099
-score FSL_HELO_FAKE 1.299
+score FSL_HELO_FAKE 1.300
score FSL_NEW_HELO_USER 0.001
-score FUZZY_AMAZON 2.199
-score FUZZY_BITCOIN 0.698
+score FUZZY_AMAZON 2.200
+score FUZZY_BITCOIN 0.714
score FUZZY_WALLET 1.899
-score GB_BITCOIN_CP 0.332
+score GB_BITCOIN_CP 0.153
score GB_CUSTOM_HTM_URI 1.499
-score GB_FAKE_RF_SHORT 0.825
+score GB_FAKE_RF_SHORT 0.826
score GB_FREEMAIL_DISPTO 0.001 # force non-zero
-score GB_HASHBL_BTC 2.441
+score GB_HASHBL_BTC 2.326
score GOOG_REDIR_NORDNS 0.147
-score GOOG_STO_IMG_HTML 3.000
+score GOOG_STO_IMG_HTML 2.999
score GOOG_STO_NOIMG_HTML 2.999
-score HAS_X_OUTGOING_SPAM_STAT 0.796
-score HDRS_LCASE_IMGONLY 0.100
-score HDRS_MISSP 2.499
+score HAS_X_OUTGOING_SPAM_STAT 0.797
+score HDRS_LCASE_IMGONLY 0.099
+score HDRS_MISSP 2.500
score HDR_ORDER_FTSDMCXX_DIRECT 0.001
score HDR_ORDER_FTSDMCXX_NORDNS 0.001
score HEADER_FROM_DIFFERENT_DOMAINS 0.250
@@ -92,54 +92,53 @@ score HK_RANDOM_FROM 0.9
score HK_RANDOM_REPLYTO 0.999
score HK_SCAM 1.999
score HOSTED_IMG_DIRECT_MX 3.499
-score HOSTED_IMG_FREEM 1.980
+score HOSTED_IMG_FREEM 1.990
score HOSTED_IMG_MULTI_PUB_01 2.999
-score HTML_BADATTR 0.400
+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 1.299
-score HTML_TEXT_INVISIBLE_STYLE 2.391
-score JH_SPAMMY_HEADERS 3.500
+score HTML_TEXT_INVISIBLE_FONT 1.313
+score HTML_TEXT_INVISIBLE_STYLE 2.390
+score JH_SPAMMY_HEADERS 3.499
score KHOP_HELO_FCRDNS 0.260
score LIST_PARTIAL_SHORT_MSG 2.499
-score LONG_HEX_URI 3.000
+score LONG_HEX_URI 2.999
score LONG_IMG_URI 2.938
-score LONG_INVISIBLE_TEXT 3.000
+score LONG_INVISIBLE_TEXT 2.999
score LOTS_OF_MONEY 0.010
score LOTTO_DEPT 1.999
-score MALFORMED_FREEMAIL 2.700
-score MALWARE_NORDNS 3.008
+score MALFORMED_FREEMAIL 2.699
+score MALWARE_NORDNS 2.100
score MALW_ATTACH 2.199
score MANY_SPAN_IN_TEXT 2.199
score MANY_SUBDOM 2.199
-score MILLION_EUR 0.539
score MILLION_HUNDRED 0.001
score MILLION_USD 1.999
score MIMEOLE_DIRECT_TO_MX 0.001
-score MIME_NO_TEXT 0.804
+score MIME_NO_TEXT 0.445
score MIXED_ES 2.299
score MIXED_HREF_CASE 1.999
-score MONEY_ATM_CARD 0.258
-score MONEY_FORM 0.001
-score MONEY_FORM_SHORT 2.393
+score MONEY_ATM_CARD 0.506
+score MONEY_FORM 0.001 # force non-zero
+score MONEY_FORM_SHORT 2.394
score MONEY_FRAUD_3 0.001
score MONEY_FRAUD_5 0.001
-score MONEY_FRAUD_8 0.297
-score MONEY_FREEMAIL_REPTO 1.412
+score MONEY_FRAUD_8 1.561
+score MONEY_FREEMAIL_REPTO 1.427
score MONEY_FROM_MISSP 0.001
-score MONEY_NOHTML 2.041
-score NA_DOLLARS 1.500
-score NICE_REPLY_A -4.020
-score NORDNS_LOW_CONTRAST 0.123
-score NO_FM_NAME_IP_HOSTN 0.295
+score MONEY_NOHTML 2.048
+score NA_DOLLARS 1.499
+score NICE_REPLY_A -3.770
+score NORDNS_LOW_CONTRAST 0.126
+score NO_FM_NAME_IP_HOSTN 0.976
score NSL_RCVD_FROM_USER 0.001
-score NSL_RCVD_HELO_USER 0.286
-score OBFU_BITCOIN 0.244
+score NSL_RCVD_HELO_USER 0.252
+score OBFU_BITCOIN 0.184
score OBFU_TEXT_ATTACH 1.599
-score ODD_FREEM_REPTO 2.595
+score ODD_FREEM_REPTO 2.594
score PDS_BAD_THREAD_QP_64 0.999
-score PDS_BTC_ID 0.500
+score PDS_BTC_ID 0.499
score PDS_BTC_MSGID 0.001
score PDS_EMPTYSUBJ_URISHRT 1.499
score PDS_FRNOM_TODOM_DBL_URL 0.001 # force non-zero
@@ -152,53 +151,51 @@ score PHP_SCRIPT 2.4
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_BODY_SINGLE_WORD 0.001
score SCC_CANSPAM_2 2.799
-score SCC_ISEMM_LID_1B 1.500
-score SENDGRID_REDIR 1.059
+score SCC_ISEMM_LID_1B 1.499
+score SENDGRID_REDIR 1.066
score SHOPIFY_IMG_NOT_RCVD_SFY 2.499
score SHORTENED_URL_SRC 2.299
score SHORT_SHORTNER 1.999
score SPOOFED_FREEMAIL 0.001
score SPOOFED_FREEMAIL_NO_RDNS 0.001
-score SPOOFED_FREEM_REPTO 1.020
-score STATIC_XPRIO_OLE 1.263
-score STOX_BOUND_090909_B 1.074
-score SUBJ_ATTENTION 0.499
-score THIS_AD 0.899
+score SPOOFED_FREEM_REPTO 0.300
+score STATIC_XPRIO_OLE 1.266
+score STOX_BOUND_090909_B 0.967
+score SUBJ_ATTENTION 0.500
+score THIS_AD 0.900
score TONOM_EQ_TOLOC_SHRT_SHRTNER 1.499
-score TO_EQ_FM_DIRECT_MX 1.895
-score TO_EQ_FM_DOM_HTML_ONLY 0.739
-score TO_EQ_FM_DOM_SPF_FAIL 0.495
-score TO_EQ_FM_HTML_ONLY 0.244
-score TO_EQ_FM_SPF_FAIL 1.082
+score TO_EQ_FM_DIRECT_MX 2.196
+score TO_EQ_FM_DOM_SPF_FAIL 1.077
+score TO_EQ_FM_SPF_FAIL 0.814
score TO_IN_SUBJ 0.099
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 2.499
+score TO_NO_BRKTS_MSFT 2.500
score TO_NO_BRKTS_NORDNS_HTML 1.999
-score TO_NO_BRKTS_PCNT 2.215
+score TO_NO_BRKTS_PCNT 2.304
score TVD_SPACE_RATIO_MINFP 0.001
-score UNDISC_FREEM 2.599
+score UNDISC_FREEM 2.500
score UNDISC_MONEY 2.799
score UNICODE_OBFU_ZW_MANY 2.999
-score URI_DOTEDU 1.185
-score URI_GOOGLE_PROXY 1.800
+score URI_DOTEDU 0.001
+score URI_GOOGLE_PROXY 1.799
score URI_IN_URI_5 3.199
-score URI_PHISH 3.999
-score URI_PHP_REDIR 2.488
+score URI_PHISH 4.000
+score URI_PHP_REDIR 2.491
score URI_TRY_3LD 1.999
score URI_WP_HACKED_2 2.499
-score VFY_ACCT_NORDNS 2.814
+score VFY_ACCT_NORDNS 2.804
score VISTA_COST 2.499
-score VISTA_TONOM_EQ_TOLOC 2.499
+score VISTA_TONOM_EQ_TOLOC 2.500
score WIKI_IMG 1.999
-score WORD_INVIS 1.903
+score WORD_INVIS 1.897
score WORD_INVIS_MANY 2.999
score XFER_LOTSA_MONEY 0.001
-score XM_DIGITS_ONLY 2.332
+score XM_DIGITS_ONLY 0.671
score XM_RANDOM 2.499
score XM_UC_ONLY 2.499
score XPRIO 0.001
@@ -292,7 +289,6 @@ score GOOG_STO_HTML_PHISH 1.0
score GOOG_STO_HTML_PHISH_MANY 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_LOTTO 1.000
@@ -385,7 +381,6 @@ score TARINGANET_IMG_NOT_RCVD_TN 1.0
score THIS_IS_ADV_SUSP_NTLD 1.000
score TONLINE_FAKE_DKIM 1.000
score TO_NAME_SUBJ_NO_RDNS 1.000
-score TRANSFORM_LIFE 1.000
score TVD_SPACE_ENCODED 1.000
score TW_GIBBERISH_MANY 1.000
score UC_GIBBERISH_OBFU 1.000

Modified: spamassassin/trunk/rulesrc/scores/stats-set0
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set0?rev=1914377&r1=1914376&r2=1914377&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set0 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set0 Wed Dec 6 02:44:44 2023
@@ -1,40 +1,40 @@
##### WITH NEW RULES AND SCORES #####

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 445582 62.205% (99.227% of non-spam corpus)
-# Correctly spam: 176822 24.685% (66.161% of spam corpus)
-# False positives: 3472 0.485% (0.773% of nonspam, 60397 weighted)
-# False negatives: 90440 12.626% (33.839% of spam, 180461 weighted)
-# Average score for spam: 7.3 nonspam: 0.6
-# Average for false-pos: 5.8 false-neg: 2.0
-# TOTAL: 716316 100.00%
+# Correctly non-spam: 447205 62.435% (99.244% of non-spam corpus)
+# Correctly spam: 175099 24.446% (65.909% of spam corpus)
+# False positives: 3406 0.476% (0.756% of nonspam, 58736 weighted)
+# False negatives: 90567 12.644% (34.091% of spam, 178726 weighted)
+# Average score for spam: 7.1 nonspam: 0.6
+# Average for false-pos: 5.9 false-neg: 2.0
+# TOTAL: 716277 100.00%

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

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 55633 99.37%
-# Correctly spam: 21559 64.51%
-# False positives: 354 0.63%
-# False negatives: 11863 35.49%
-# TCR(l=50): 1.130535 SpamRecall: 64.505% SpamPrec: 98.385%
+# Correctly non-spam: 55753 99.21%
+# Correctly spam: 21796 65.59%
+# False positives: 443 0.79%
+# False negatives: 11437 34.41%
+# TCR(l=50): 0.989460 SpamRecall: 65.585% SpamPrec: 98.008%

##### 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: 437424 97.41%
-# Correctly spam: 114550 42.86%
-# False positives: 11630 2.59%
-# False negatives: 152712 57.14%
-# TCR(l=50): 0.364012 SpamRecall: 42.861% SpamPrec: 90.783%
+# Correctly non-spam: 438920 97.41%
+# Correctly spam: 113974 42.90%
+# False positives: 11691 2.59%
+# False negatives: 151692 57.10%
+# TCR(l=50): 0.360841 SpamRecall: 42.901% SpamPrec: 90.697%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 54507 97.36%
-# Correctly spam: 14324 42.86%
-# False positives: 1480 2.64%
-# False negatives: 19098 57.14%
-# TCR(l=50): 0.358998 SpamRecall: 42.858% SpamPrec: 90.635%
+# Correctly non-spam: 54751 97.43%
+# Correctly spam: 14116 42.48%
+# False positives: 1445 2.57%
+# False negatives: 19117 57.52%
+# TCR(l=50): 0.363731 SpamRecall: 42.476% SpamPrec: 90.714%

Modified: spamassassin/trunk/rulesrc/scores/stats-set1
URL: http://svn.apache.org/viewvc/spamassassin/trunk/rulesrc/scores/stats-set1?rev=1914377&r1=1914376&r2=1914377&view=diff
==============================================================================
--- spamassassin/trunk/rulesrc/scores/stats-set1 (original)
+++ spamassassin/trunk/rulesrc/scores/stats-set1 Wed Dec 6 02:44:44 2023
@@ -1,23 +1,23 @@
##### WITH NEW RULES AND SCORES #####

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 421488 85.370% (99.328% of non-spam corpus)
-# Correctly spam: 27310 5.531% (39.363% of spam corpus)
-# False positives: 2851 0.577% (0.672% of nonspam, 31796 weighted)
-# False negatives: 42069 8.521% (60.637% of spam, 60410 weighted)
+# Correctly non-spam: 421473 85.367% (99.325% of non-spam corpus)
+# Correctly spam: 27307 5.531% (39.360% of spam corpus)
+# False positives: 2866 0.580% (0.675% of nonspam, 31920 weighted)
+# False negatives: 42071 8.521% (60.640% of spam, 60354 weighted)
# Average score for spam: 6.0 nonspam: -1.0
# Average for false-pos: 6.8 false-neg: 1.4
-# TOTAL: 493718 100.00%
+# TOTAL: 493717 100.00%

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

# SUMMARY for threshold 5.0:
-# Correctly non-spam: 52577 99.42%
-# Correctly spam: 3357 38.32%
-# False positives: 307 0.58%
-# False negatives: 5404 61.68%
-# TCR(l=50): 0.422135 SpamRecall: 38.318% SpamPrec: 91.621%
+# Correctly non-spam: 52578 99.42%
+# Correctly spam: 3354 38.29%
+# False positives: 306 0.58%
+# False negatives: 5406 61.71%
+# TCR(l=50): 0.423066 SpamRecall: 38.288% SpamPrec: 91.639%

##### 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: 422487 99.56%
-# Correctly spam: 14942 21.54%
+# Correctly spam: 14948 21.55%
# False positives: 1852 0.44%
-# False negatives: 54437 78.46%
-# TCR(l=50): 0.471847 SpamRecall: 21.537% SpamPrec: 88.972%
+# False negatives: 54430 78.45%
+# TCR(l=50): 0.471863 SpamRecall: 21.546% SpamPrec: 88.976%
Reading scores from "../rules-base"...
Reading per-message hit stat logs and scores...

# SUMMARY for threshold 5.0:
# Correctly non-spam: 52673 99.60%
-# Correctly spam: 1899 21.68%
+# Correctly spam: 1887 21.54%
# False positives: 211 0.40%
-# False negatives: 6862 78.32%
-# TCR(l=50): 0.503159 SpamRecall: 21.676% SpamPrec: 90.000%
+# False negatives: 6873 78.46%
+# TCR(l=50): 0.502784 SpamRecall: 21.541% SpamPrec: 89.943%