Parent Directory | Revision Log | Patch
--- spamassassin/trunk/spamd/README 2005/08/18 02:28:56 233281 +++ spamassassin/trunk/spamd/README 2005/08/18 03:03:01 233282 @@ -112,11 +112,13 @@ you will need to either 2. Alternatively, let the users train their individual Bayes database. -We have implemented an auto-learning algorithm (option -'bayes_auto_learn', on by default) which can use high-scoring -and low-scoring (options 'bayes_auto_learn_threshold_spam' and -'bayes_auto_learn_threshold_nonspam') mails to improve classification -efficiency. +http://wiki.apache.org/spamassassin/SiteWideBayesFeedback can be very +helpful here. + +We have implemented an auto-learning algorithm (option 'bayes_auto_learn', on +by default) which can use high-scoring and low-scoring (options +'bayes_auto_learn_threshold_spam' and 'bayes_auto_learn_threshold_nonspam') +mails to improve classification efficiency. Security And User-Spoofing Clients
infrastructure at apache.org | ViewVC Help |
Powered by ViewVC 1.1.26 |