Skip to content

Instantly share code, notes, and snippets.

@cpragadeesh
Created July 24, 2017 19:27
Show Gist options
  • Save cpragadeesh/2db070ccff16ca5e366674697b9a682f to your computer and use it in GitHub Desktop.
Save cpragadeesh/2db070ccff16ca5e366674697b9a682f to your computer and use it in GitHub Desktop.
sample rescore results
epoch: 99 | error: 0.0519916767862 4
Statistics:
Time taken: 9.25s
Pre-rescore test data stats:
Accuracy: 69.85 %
F-score: 0.577464788732
Post-rescore test data stats:
Accuracy: 78.89 %
F-score: 0.761363636364
Optimal spam threshold: 19
SYMBOL OLD SCORE NEW SCORE
FAKE_REPLY 1.0 1.59
URI_COUNT_ODD 1.0 1.59
INVALID_FROM_8BIT 6.0 9.53
MID_RHS_WWW 0.5 0.8
BROKEN_HEADERS 10.0 15.88
HAS_X_PRIO_FIVE 0.0 0.0
TO_EXCESS_BASE64 1.5 2.38
MID_RHS_IP_LITERAL 0.5 0.8
HEADER_FORGED_MDN 2.0 3.18
MIME_HTML_ONLY 0.2 0.32
SUBJECT_ENDS_QUESTION 1.0 1.59
FROM_EXCESS_QP 1.2 1.91
MID_RHS_MATCH_FROM 0 0.0
MIME_BASE64_TEXT 0.0 0.0
TO_EQ_FROM 0.0 0.0
FROM_EQ_ENVFROM 0.0 0.0
RCVD_COUNT_ONE 0.0 0.0
TO_DN_NONE 0.0 0.0
R_BAD_CTE_7BIT 4.0 6.35
TO_EXCESS_QP 1.2 1.91
R_MIXED_CHARSET 5.0 7.94
HAS_XOIP 0.0 0.0
SUBJECT_NEEDS_ENCODING 1.0 1.59
FROM_HAS_DN 0.0 0.0
RCPT_COUNT_TWO 0.0 0.0
R_MISSING_CHARSET 2.5 3.97
MID_CONTAINS_FROM 1.0 1.59
TAGGED_RCPT 0.0 0.0
HAS_X_PRIO_ONE 0.0 0.0
MISSING_DATE 1.0 1.59
MIME_MA_MISSING_HTML 1.0 1.59
MIME_GOOD -0.1 -0.16
MID_BARE_IP 2.0 3.18
FORGED_OUTLOOK_HTML 5.0 7.94
MIME_HEADER_CTYPE_ONLY 2.0 3.18
MISSING_SUBJECT 2.0 3.18
FORGED_MUA_OUTLOOK 3.0 4.77
INTRODUCTION 2.0 3.18
TO_DN_SOME 0.0 0.0
MISSING_FROM 2.0 3.18
RCVD_COUNT_FIVE 0.0 0.0
HTTP_TO_IP 1.0 1.59
FROM_NO_DN 0.0 0.0
DATE_IN_PAST 1.0 1.59
MIME_UNKNOWN 0.1 0.16
FAKE_REPLY_C 6.0 9.53
CT_EXTRA_SEMI 1.0 1.59
HAS_X_PRIO_THREE 0.0 0.0
FROM_NAME_EXCESS_SPACE 1.0 1.59
HFILTER_HOSTNAME_4 2.5 3.97
MIME_BAD_ATTACHMENT 4.0 6.35
RCPT_COUNT_THREE 0.0 0.0
PREVIOUSLY_DELIVERED 0.0 0.0
PHISHING 4.0 6.35
R_PARTS_DIFFER 1.0 1.59
HAS_XAW 0.0 0.0
RCVD_TLS_ALL 0.0 0.0
MISSING_MIMEOLE 2.0 3.18
SUSPICIOUS_RECIPS 1.5 2.38
EXT_CSS 1.0 1.59
TO_DN_EQ_ADDR_ALL 0.0 0.0
TO_DN_ALL 0.0 0.0
TO_DN_EQ_ADDR_SOME 0.0 0.0
HAS_WP_URI 0.0 0.0
R_UNDISC_RCPT 3.0 4.77
RCVD_COUNT_ZERO 0.0 0.0
TO_MATCH_ENVRCPT_ALL 0.0 0.0
ONCE_RECEIVED 0.1 0.16
R_SUSPICIOUS_IMAGES 5.0 7.94
DMARC_NA 0 0.0
MV_CASE 0.5 0.8
HAS_ORG_HEADER 0.0 0.0
PRECEDENCE_BULK 0.0 0.0
RCPT_COUNT_ZERO 0.0 0.0
MID_MISSING_BRACKETS 0.5 0.8
MISSING_MIME_VERSION 2.0 3.18
R_SPF_DNSFAIL 0.0 0.0
MISSING_MID 2.5 3.97
URL_IN_SUBJECT 4.0 6.35
SUBJECT_HAS_EXCLAIM 0.0 0.0
INVALID_MSGID 1.7 2.7
RCPT_COUNT_SEVEN 0.0 0.0
HAS_ATTACHMENT 0 0.0
RCVD_COUNT_TWO 0.0 0.0
BROKEN_CONTENT_TYPE 1.5 2.38
FROM_NAME_HAS_TITLE 1.0 1.59
SUBJ_ALL_CAPS 3.0 4.77
MIME_MA_MISSING_TEXT 2.0 3.18
RCPT_COUNT_FIVE 0.0 0.0
SUBJECT_ENDS_SPACES 0.5 0.8
R_DKIM_NA 0 0.0
RCPT_COUNT_ONE 0.0 0.0
FORGED_MUA_MAILLIST 0.0 0.0
FROM_NEQ_ENVFROM 0.0 0.0
HTML_SHORT_LINK_IMG_3 0.5 0.8
HTML_SHORT_LINK_IMG_2 1.0 1.59
HTML_SHORT_LINK_IMG_1 2.0 3.18
TO_DOM_EQ_FROM_DOM 0.0 0.0
RCVD_COUNT_THREE 0.0 0.0
MID_RHS_NOT_FQDN 0.5 0.8
SUBJECT_ENDS_EXCLAIM 0.0 0.0
FROM_EXCESS_BASE64 1.5 2.38
CTYPE_MIXED_BOGUS 0.1 0.16
FORGED_MUA_THEBAT_MSGID_UNKNOWN 3.0 4.77
REPTO_QUOTE_YAHOO 2.0 3.18
SUBJECT_HAS_CURRENCY 1.0 1.59
MAILLIST -0.2 -0.32
RATWARE_MS_HASH 2.0 3.18
MISSING_TO 2.0 3.18
FORGED_MUA_KMAIL_MSGID 3.0 4.77
TAGGED_FROM 0.0 0.0
HAS_X_ANTIABUSE 0.0 0.0
INVALID_RCPT_8BIT 6.0 9.53
DATE_IN_FUTURE 4.0 6.35
RCVD_COUNT_SEVEN 0.0 0.0
RCPT_COUNT_TWELVE 0.0 0.0
RCVD_COUNT_TWELVE 0.0 0.0
SUBJECT_HAS_QUESTION 0.0 0.0
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment