Fritz Borgstedt wrote:
This way of handling the "corpus" shows a misunderstanding of the bayes concept. This manually manipulating is contrary to the idea of ASSP. Fritz, how is this a misunderstanding of the Bayesian concept? As I understand it, the Bayesian corpus is as good as what you feed it. If you have whitelisted users polluting it with bad mail (and by bad mail I mean utter crap), then that only weakens your corpus to spam with similar words. Words like profanity, references to purchases, credit cards, etc,etc... References that (depending on the volume) weaken the spam values in the corpus. In *my* organization, eliminating corpus pollution is a priority for me. A large percentage of my users are very abusive of the corporate mail system - and my management will not enforce policy. I see nothing wrong with massaging the corpus. I know it has been a benefit to me in terms of false-positives and negatives. But of course YMMV... |
------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________ Assp-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/assp-user
