Hi Junkmail-users, We've prepared a website where you can read more about SpamChk. http://www.riedmann.it/spamchk/ There is a distribution archive, a installation guide for declude-users and a history.
Some additional notes and recommendations: The tests, actions and keywords are well tested. Any single test has a reason based on statistic analysis ond weeks of research on thousands of e-mail messages. A longer list of keywords can increase the detection rate, but then you should also add some negative keywords - specially used in auto-generated emails like newsletters, wheater reports, booking requests and confirmations. This can differ from system to system and country to country. We're _very_ happy if there is someone making a research on tipical keywords and phrases used in body, subject, x-mailer, ... on his system. The results can help to all other users of SpamChk. (The lists posted by Kami some days ago are very usefull) If you have any suggestion on the internal weighting system or the default ini-settigns of SpamChk we ask only to make out a case why. (not based only on a single case) For programming errors and suggestions you can send a message to [EMAIL PROTECTED] Wolfgang Riedmann is the mind and programmer behind this project. As SpamChk can also return negative error codes we appreciate also suggestions for tests that triggers tipical non-spam messages. With this strategy we not only want to classify messages between "neutral" (0) and "spam" (+x) but between "not spam" (-x) and "spam" (+x) Markus --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type "unsubscribe Declude.JunkMail". The archives can be found at http://www.mail-archive.com.
