URL Protection - Machine Learning Detection - Jul 2022

Service Update

Availability July 8th, 2022
Product(s) Email Security Cloud Gateway (CG)
Who's affected TTP URL Protection users

Overview

Mimecast is pleased to announce the release of a new machine learning model that enhances resilience against zero-day and emerging phishing attacks.  This is a feature-rich machine learning model for URL classification, designed to target zero-day and emerging threats with a low false-positive rate.

What's changing

The model can analyze URLs when clicked as well as via the Check URL function in the Administration Console. Detections made by the model will be shown in the Targeted Threat Protection (TTP) URL Logs – Scan details section in the Mimecast Administration Console


image.png

The model is trained on a high-confidence data set comprised of 1 million samples. With features engineered from a wide variety of elements, such as lexical URL analysis, page content, domain analysis, certificate details, and email content, the model has context beyond the URL alone.

Recommended actions 

The machine learning model is fully enabled for customers with TTP URL Protect set to an Aggressive categorization mode.

Action Required: The URL Category Scanning drop-down must be set to aggressive in the URL Protection definition in order to take advantage of this feature.

image.png 

The Rewrite Mode does not have to be set to Aggressive.
For more information on Rewrite Modes, see Configuring URL Protection Definitions Inbound Settings section.

See Also...

Was this article helpful?
0 out of 0 found this helpful

Comments

0 comments

Please sign in to leave a comment.