How indentification Spam by Kapersky Lab

We all know that spam is annoying and harmful. But not all of us know how to identify junk e-mail is unwanted. As a result we may be stuck clicking a malicious link.

One of the most effective way deter spam by security vendors is to classify messages based on keywords and phrases that characterize spam. In this way the system can quickly configure to block new types of spam with false positive detection rate is low.

Well, in the framework of how to identify spam, on 16 November 2010 developer content security and threat management solutions Kaspersky Lab received a patent U.S. Patent and Trademark Office of the United States number 7,836,061 for its technology that uses keywords or phrases.

Describe the method patented by Kaspersky Lab? Look. Classified electronic text messages on a hierarchical list of categories of messages, known as rubricator. Each incoming text messages categorized by calculating the relative amounts for each category that contains the key terms in the message, and then determine the level of similarity of each template. If a text message containing a number of keywords, or quite similar to the one specified templates, and fits into one category, the message including spam.

This method also allows the creation of categories of messages manually through a key word or template. Each category can be grouped into subcategories for classification is more appropriate. To be more simple categories, text messages can be identified prior to use certain techniques, such as automatic language detection, delete the words that are often used, such as articles or prepositions, as well as filter out any interference.