Appendix F. DSPAM

DSPAM (as in de-spam, or to eliminate spam) is a high-performance and highly accurate Bayesian-style classifier designed for large installations. It contains many of the features present in other leading Bayesian classifiers, including

  • Per-user spam corpus

  • Easy filter retraining

  • Graphical user interface

  • Proxy/MDA masquerade support for easy installation

  • Integration with all major MTAs

  • Support for all users or per-user integration

It also has features that are relatively unique to DSPAM:

  • Designed expressly for large-scale deployments and efficiency

  • Can be configured to be run as a POP3 proxy, making installation/configuration easier

  • Supports multiple types of back-end databases for storing data, including

    • SQLite

    • Berkeley DB3

    • Berkeley DB4 ...

Get Slamming Spam: A Guide for System Administrators now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.