I am also seeing ham heavy corpus....what settings should I use to try and get it back to a good level in autocorrectCorpus?
May-01-12 00:38:26 Spam Weight: 357,445 May-01-12 00:38:26 Not-Spam Weight: 3,369,188 May-01-12 00:38:26 Corpus norm: 0.1061 - (warning: extremely ham heavy) May-01-12 00:38:26 Corpus confidence: 0.07772573 May-01-12 00:38:26 Recommendation: RebuildSpamDB will limit the number of used messages in your corpus. Excess files will be ingored. May-01-12 00:38:26 Corpus norm should be between 0.6 and 1.4 -----Original Message----- From: Thomas Eckardt [mailto:[email protected]] Sent: Tuesday, May 01, 2012 7:07 AM To: ASSP development mailing list Subject: [Assp-test] Antwort: Ham-heavy corpus You can use the 'autocorrectCorpus' option - or 'MaxBayesFileAge' to remove some files from the notspam folder. Thomas Von: Scott MacLean <[email protected]> An: ASSP development mailing list <[email protected]> Datum: 30.04.2012 22:25 Betreff: [Assp-test] Ham-heavy corpus My corpus norm has gradually crept up from about 0.2 to its present 0.53 - and never goes any higher. I have it set for 15000 messages both in spam and notspam, there are 14820 currently in spam and 13653 in notspam. What else do I need to do to get my corpus norm up higher? Apr-30-12 12:30:54 Corpus norm: 0.5294 - (warning: extremely ham heavy) Apr-30-12 12:30:54 Corpus confidence: 0.21379828 Apr-30-12 12:30:54 Corpus norm should be between 0.6 and 1.4 Apr-30-12 12:30:54 Recommendation: You need more spam messages in the corpus. Apr-30-12 12:30:54 starting auto correction for corpus - delete old ham files from notspam Apr-30-12 12:30:58 info: starting cleanup for to much (old) files in folder D:/ASSP/notspam info: deleted 26 old files from folder D:/ASSP/notspam ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Assp-test mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/assp-test DISCLAIMER: ******************************************************* This email and any files transmitted with it may be confidential, legally privileged and protected in law and are intended solely for the use of the individual to whom it is addressed. This email was multiple times scanned for viruses. There should be no known virus in this email! ******************************************************* ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Assp-test mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/assp-test
