Introduction

AntispamSniper for The Bat! provides professional anti-spam and anti-phishing protection for your mailbox. A combination of several methods for automatic email classification results in an excellent filtering quality with the minimum error rate. The plug-in also has a built-in option allowing you to delete spam on the server by message headers. Good messages deleted by mistake can be restored within a certain period of time after deletion. The plug-in filters POP3 and IMAP accounts (filtering IMAP is not available in free version).

What methods are used?

First of all, it is a statistical trainable classification algorithm with unique characteristics. It analyzes your private correspondence, singles out distinguishing features from messages of different classes and uses this information for the effective analysis of incoming messages. The algorithm constantly improves its classification quality with each new processed message.
Messages from whitelisted senders are not filtered by other methods and are classified as non-spam. The whitelist can be optionally updated automatically.
Messages received from blacklisted senders are classified as spam. The blacklist is optionally updated automatically.
Filtering rules detect spam and normal messages by specific combinations of characters in their bodies and headers. These rules are optional because the trained algorithm and the whitelist are very effective. But carefully compiled lists of rules may improve the classification quality in those cases when the algorithm cannot select a certain class for a message with high probability.
Exceptions are empirical filtering conditions. For example, large messages are probably non-spam. The presence of certain types of attached files can also be a good sign to consider a message as non-spam.
The anti-phishing filter blocks messages with spoof links and thus protects you against phishing attacks.
(pro) DNSBL services are used to block spam by the IP address of the sender.
(pro) URIBL services are used to block spam by the domains in message text.
(pro) The list of black words is used to delete spam by keywords in message subjects.
(pro) The plug-in optionally learns outgoing messages as non-spam.
(pro) The plug-in can delete spam on the server using the list of accounts. These filtering accounts can be added either manually or automatically (in case of non-secure connections).
Messages can be blocked by attachments types. The plug-in automatically detects automatically attached Windows executable modules, animated GIF images and images in HTML linked to fraudulent servers.

The classifier starts working right after the installation and detects spam with a quite high quality. To improve the filtering quality, you should configure the plug-in according to instructions from Getting started.