Machine Learning Power for Messaging Security

The power behind F-Secure's enterprise messaging security solutions - Proofpoint MLX - is an advanced message classification system developed by the scientists at Proofpoint's Anti-Spam Laboratory. Based on advanced statistical techniques including logistic regression and support vector machines, MLX enables the accurate classification and identification of unstructured content, such as the contents of e-mail messages and valuable company documents.

Most Accurate and Confident Anti-Spam Technology

MLX is the basis for the unrivalled anti-spam accuracy delivered by the F-Secure Messaging Security Gateway appliance. Using MLX's machine learning capabilities, the Spam Detection module analyzes more than 100,000 structural and content attributes to accurately differentiate between spam and valid messages. Traditional anti-spam solutions evaluate only a limited number of attributes and are unable to decisively classify spam, leading to low effectiveness and a high rate of false positives. MLX is far superior to simple statistical techniques such as Bayesian filters-and it doesn't rely on signatures or fingerprinting techniques, which are easily fooled by spammers.

Click here to view a Flash movie that shows how MLX detects and eliminates spam. This movie illustrates how MLX uses advanced machine learning techniques - such as support vector machines and logistic regression - to analyze more than 100,000 attributes in every e-mail, providing extreme confidence in identifying spam messages.

The highly-confident spam scoring provided by MLX lets you take decisive action against spam messages-deleting them before they clog your internal mail servers and vastly reducing the number of "possible spam" messages that need to be quarantined. The screenshot below illustrates the high confidence with which MLX scores spam. Note the strong clustering of scores near 0 (definitely not spam) and 100 (definitely spam):

Ongoing Training for Futureproof Protection

Intelligent anti-spam technology is continually being trained by Proofpoint scientists to defend against new forms of spam. This training allows MLX to predict and adapt to new forms of spam as they appear. Unlike other anti-spam solutions, Proofpoint's ability to defend against spam attacks does not degrade over time-and updates to the MLX anti-spam engine are automatically delivered to your servers on a regular basis. MLX is constantly evolving to counter emerging threats, ensuring that your messaging infrastructure is secure against tomorrow's spammers as well as today's.

Machine Learning Applications Beyond Spam

Proofpoint scientists and engineers continue to apply MLX machine learning and advanced statistical technologies to meet new messaging security challenges. MLX technology has many more applications beyond spam detection. In latest products, MLX powers the advanced content security features of the Proofpoint Digital Asset Security module and the intelligent perimeter security features of E-mail Firewall and MLX Dynamic Reputation service.

For example, in the Proofpoint Digital Asset Security module, MLX technology has now been applied to the onsite analysis and classification of specific documents. Proofpoint Digital Asset Security uses MLX technology to create a statistical representation of all trained documents, and then compares all messages against these statistical representations, detecting a "match" if even a small segment of the confidential content is being communicated.

Continued Research and Development Protects Your Investment

Proofpoint is the only vendor to apply these powerful machine learning techniques to messaging security. Scientists at Proofpoint's Anti-Spam Laboratory continue to conduct primary research into new, advanced statistical techniques and to develop new defences based on MLX. This ongoing research and development ensures that solutions are always one step ahead of threats to the security of your messaging infrastructure.