Bayesian filters are adaptive and become more efficient at correctly identifying spam by keeping the filters trained. Training requires one to feed the filter with known spam and known ham (valid emails) so that it learns the characteristics of both types of email.
Trained by Administrator
The administrator feeds the filter with known spam and known ham (valid emails) so that it learns the characteristics of both types of email. Requires the administrator to accurately determine what emails are spam and what emails aren't spam before training the Bayesian filter.
Trained by Users
The users feed the filter with known spam and known ham (valid emails) so that it learns the characteristics of both types of email. Requires the users to accurately determine what emails are spam and what emails aren't spam before training the Bayesian filter.
Keyword filters attempt to identify spam by searching for specific words within emails. i.e. words that are frequently found in spam but not in ham (valid emails).
Pre-Defined Keyword List
A Pre-Defined Keyword List is compiled by the vendor of the antispam solution and supplied to its customers.
Customizable Keyword List
A Customizable Keyword List allows administrators to add and remove keywords from the list for improved accuracy.
Boolean Logic allows your keyword list to include boolean operators such as AND, OR, NOR, NOT etc, thereby improving the accuracy of the Keyword Filter. For example: Classify email as spam if and only if the keywords "lowest" AND "mortgage" are found.
Blacklist filters attempt to reduce spam by blocking all emails from a sender or group of senders that you, your organization or the public has determined to be spammers.
Block specific IP Addresses
Permits blocking of all emails received from a specific IP address.
Block specific Server
Permits blocking of all emails received from a specific mail server.
Block email addresses
Permits blocking of all emails received from a specific email address.
Public Blacklist support
Uses a blacklist/s compiled by the antispam community to block spam.
Whitelist filters attempt to reduce false positives by allowing all emails from a sender or group of senders that you or your organization has determined to be a non-spammer.
Allow specific IP addresses
Permits delivery of all emails received from a specific IP address.
Allow specific server
Permits delivery of all emails received from a specific mail server.
Allow email addresses
Permits delivery of all emails received from a specific email address.
Filters that classify and identify spam based on content. For example, messages may be categorized as "Sexually Explicit", "Get Rich Quick Schemes" etc., while also allowing individual spam threshold or sensitivity settings to be applied to each category.
Pre-Set Categories are defined by the vendor of the antispam solution.
User Defined Categories
Allows an organization to define categories based on message content that may be unique to that organization's incoming email.
Signature Filters identify spam based on the specific attributes of known spam messages and block any new messages that have attributes that are identical to known spam messages.
Automatically downloads new signatures periodically from the antispam vendor to successfully identify the latest spam messages.
Header Analysis Filter
An email's header contains a complete list of the path traveled by the email along various machines on the Internet before reaching its destination. Header analysis is used to determine whether the header has been tampered with, which is evidence that the message may be spam.
Character Set Filter
Permits blocking of all emails that use a specified character set or sets. Particularly useful in blocking some, but not all, foreign language spam.
An authentication method used in which an unrecognized sender is prompted to provide unique information. The message will be delivered to the recipient only if the information provided is accurate.
An area in which suspicious messages are stored until they are reviewed by the recipient or administrator to determine what further action will be taken.
Accessible to Users
Recipients may review messages which are stored in the quarantine.
The solution is able to perform one or more tasks once a message is identified as spam.
Bounce "Error" Message
Sends a fake "error" message to the sender of a spam message, thereby leading the spammer to believe that the recipient's email address is invalid. This is used to encourage the removal of the address from the spammer's mailing list to prevent further such messages from being sent.
The solution immediately deletes any messages identified as spam.
A set of criteria that determines whether or not a message is spam.
Rule Creation Customization
Allows the administrator to define the criteria that will determine whether or not a message is spam.
The solution is shipped with a set of rules that are defined by the vendor.
Multiple Rule Selection
Allows the administrator to select multiple rules for enhanced effectiveness in determining whether or not a message is spam.
Sensitivity settings allow the administrator to adjust the aggressiveness of the filtering performed so that an acceptable balance between the number false-positives and the number of false-negatives may be found.
Allows the administrator to define the criteria that will determine the aggressiveness of the filtering performed.
Individual User Profiles
The solution offers personalized filtering performance and configurations for each user within the organization.
The solution offers distinct filtering performance and configurations for each group within the organization.
Solution provides reports and statistics on email activity.
Open Relay Protection
Prevents spammers from using your organization's mail server to deliver spam messages to a destination outside the network.
Spoofing is the practice of forging a return email address. Anti-spoofing is used to reject messages that have been sent from outside the network yet claim to have been sent by someone from within the network.
Spam messages that are sent in HTML format may contain hidden code that enables the sender to verify the validity of the recipient's address once the message is opened. HTML stripping prevents such verification by delivering the email in plain text.
Many solutions may be deployed within a variety of different email/messaging environments but not necessarily within all environments. Your existing environment and the possibility of upgrading or migrating to a new environment are factors to consider when choosing a solution.
Domino 6.x supported.
Domino 5.x supported.
Domino 4.x supported.
Exchange 2003 supported.
Exchange 2000 supported.
Exchange 5.5 supported.
GroupWise 6.x supported.
GroupWise 5.x supported.
Supports one or more flavors of Unix and/or Linux.