Deep Generative Models to Extend Active Directory Graphs with Honeypot Users

TL;DR

This paper introduces a novel machine learning approach using Variational Autoencoders and Bidirectional DAG-RNNs to generate strategically placed honeyusers in Active Directory graphs, enhancing attack detection by luring intruders.

Contribution

It presents a new method combining VAEs and DAG-RNNs to automatically generate effective honeyusers in AD structures, improving cybersecurity defenses.

Findings

Generated honeyusers are well-positioned to attract attackers.

The model's AD structures closely resemble original graphs.

Intruders are successfully lured into honeyusers in experiments.

Abstract

Active Directory (AD) is a crucial element of large organizations, given its central role in managing access to resources. Since AD is used by all users in the organization, it is hard to detect attackers. We propose to generate and place fake users (honeyusers) in AD structures to help detect attacks. However, not any honeyuser will attract attackers. Our method generates honeyusers with a Variational Autoencoder that enriches the AD structure with well-positioned honeyusers. It first learns the embeddings of the original nodes and edges in the AD, then it uses a modified Bidirectional DAG-RNN to encode the parameters of the probability distribution of the latent space of node representations. Finally, it samples nodes from this distribution and uses an MLP to decide where the nodes are connected. The model was evaluated by the similarity of the generated AD with the original, by the positions of the new nodes, by the similarity with GraphRNN and finally by making real intruders attack the generated AD structure to see if they select the honeyusers. Results show that our machine learning model is good enough to generate well-placed honeyusers for existing AD structures so that intruders are lured into them.