Automatic uniform resource locator-based message filter
AssigneeAT&T Corp.
InventorBill C. Riemers
Life of this patent
Estimating the term…
Abstract
A system and method for filtering messages to identify spam. A Uniform Resource Locator (URL) is identified in a message, and information is fetched from the site designated by the URL. The fetched information is analyzed to determine if the message that includes the URL is spam. If the message is spam, then a message filtering action is performed, such as deleting the message, displaying the message to the recipient with a flag, and sending the message to a third party.
Claims43
A method for determining if an electronic message is spam including:
- receiving a message;
- identifying a URL included in the message;
- fetching information corresponding to the identified URL;
- and
- determining the likelihood that the message is spam based upon an analysis of the fetched information.
An apparatus for determining if an electronic message is spam, comprising:
- a processor;
- and
- a memory storing filtering instructions adapted to be executed by said processor to identify a URL in a message, fetch information corresponding to the URL, and automatically analyze the information to determine if the message in which the URL is included is likely to be spam, said memory coupled to said processor.
A medium storing instructions adapted to be executed by a processor to perform steps including:
- receiving a message;
- identifying a URL included in the message;
- fetching information corresponding to the identified URL;
- determining a message score based upon an analysis of the fetched information, wherein the message score relates the likelihood that the message is spam.
A system for determining if an electronic message is spam, comprising:
- a. means for automatically identifying a URL included in a received message;
- b. means for fetching information corresponding to the identified URL;
- and
- c. means for determining if the message containing the URL is likely to be spam.
Dependencies are inferred from the claim text.