DSPAM is a server-side anti-spam agent for UNIX email servers. It masquerades as the email server's local delivery agent and filters/learns SPAM using a Bayesian statistical approach which provides an administratively maintenance-free, self-learning anti-spam service. Each email is broken down into its most interesting tokens, each assigned a spam probability. All probabilities are then combined to produce a statistical probability of spam. This approach, applied to a mature corpus of email, has the potential to yield a 99.5% success rate with only 0.03% chance of false positives.
released on 30 November 2005
|License||Verified by||Verified on||Notes|
|GPLv2orlater||Janet Casey||29 April 2003|
Leaders and contributors
|Jonathan A. Zdziarski||Maintainer|
Resources and communication
|Required to use||Berkeley DB|
|Weak prerequisite||PERL (for optional CGI interface|
This entry (in part or in whole) was last reviewed on 30 November 2005.