A technology is in development that more realistically emulates user activities to improve cyber testbeds and ultimately prevent cyberattacks.

The Data Driven User Emulation (D2U), uses machine learning to simulate actual users’ actions in a network and then enhances cyber analysts’ ability to thwart, expose and mitigate network vulnerabilities.

Whether testing intrusion detection systems, conducting training exercises, or creating data sets to be used by the broader cybersecurity community, realistic user behavior is a critical component of a cyber range.

Existing methods either rely on network level data or replay recorded user actions to approximate real users in a network.

This technology produces generative models trained on actual user data (sequences of application usage) collected from endpoints. Once trained to the user’s behavioral data, these models can generate novel sequences of actions from the same distribution as the training data.

Schneider Bold

These sequences of actions are then fed to custom software via configuration files, which replicate those behaviors on end devices. Notably, the models are platform agnostic and could generate behavior data for any emulation software package.

“Understanding and modeling individual user behaviors is critical for cybersecurity,” said Oak Ridge National Laboratory’s (ORNL) Sean Oesch. “D2U can create unlimited, realistic test users of a particular network for developers of cyber tools to improve their products.”

Where other user models need large numbers of testers or make assumptions about their behavior, D2U needs a small number of users and emulates actual user behavior.

The software is currently deployed to help evaluate defensive cyber technologies but could have benefits to the broader cyber community.

Click here for more information.

The D2U model categorizes user data by capturing behavior in all open programs throughout a user’s day.
Source: Nathan Armistead/ORNL

Pin It on Pinterest

Share This