Kaspersky Lab patents an advanced spam identification method

25 Nov 2010
Spam News

Kaspersky Lab, a leading developer of secure content and threat management solutions, announces that it has received a US patent for an innovative technology which classifies electronic text messages for the purpose of identifying spam.

Spam causes significant damage to business and home users. Unwanted emails often contain fraudulent offers, malicious attachments or links to malicious websites.

One of the most popular and effective methods of combating unwanted emails involves classifying messages based on whether they contain key words and phrases which are typical of spam. This approach can be used to quickly configure the system to block new types of spam. It also provides high detection rates with minimal false positives.

The new technology patented by Kaspersky Lab is based on the abovementioned system of using key words and phrases to detect spam. The new technology was registered by the United States Patent and Trademark Office on 16 November, 2010, as patent number 7,836,061.

The patented method classifies electronic text messages based on a hierarchical list of message categories known as a rubricator. Each category is defined by a set of key terms and a number of text templates. An incoming text message is categorized by calculating its weighting relative to each category that contains key terms present in the message, and then determining the degree of similarity to each of the templates. If a text message includes a certain number of key words, or is sufficiently similar to one of the templates, it is matched to one of the categories, including spam.

It is possible to manually create message categories using key terms and templates. Each category can be divided into subcategories in order to provide more precise classification. To further improve and simplify categorization, text messages can be preprocessed using such techniques as automatic language detection, removing frequently used words such as articles and prepositions, etc, and filtering out noise.

"Spam is one of the most irritating and disagreeable threats, and any deterioration in the quality of filtration immediately becomes obvious to everybody. This is why it is essential for us to maintain the high quality of protection provided by the anti-spam components of our products and continually develop new technologies," said Andrey Nikishin, Director, Cloud & Content Technologies Research. "We are glad that another important company achievement in the area of anti-spam technology has received appropriate recognition in the form of a patent."

Patent authorities in the US, Russia, China and Europe are currently examining over seventy patent applications from Kaspersky Lab that cover unique and innovative technologies in the field of information security.