Computer Science > Social and Information Networks
[Submitted on 15 Jan 2024]
Title:Multi-Task DNS Security Analysis via High-Order Heterogeneous Graph Embedding
View PDF HTML (experimental)Abstract:DNS is an essential Internet infrastructure to support network applications and services, but is also a significant tool exploited by various cyberattacks. Existing DNS security analysis techniques mostly focus on one specific task associated with one single entity (e.g., domain) via conventional feature engineering. They rely heavily on the labor-intensive feature selection and largely ignore the intrinsic correlations among the heterogeneous DNS entities (e.g., domain and IP). In this paper, I explore the potential of heterogeneous graph embedding to automatically learn the behavior features of multiple DNS entities, and to simultaneously support more than one security tasks. Considering the joint optimization of malicious domain detection and IP reputation evaluation as an example, I propose a novel joint DNS embedding (JDE) model to formulate the DNS query behavior via a similarity-enhanced graph with heterogeneous entities. The random walk technique is applied to the heterogeneous graph to comprehensively explore the hidden homogeneous and heterogeneous high-order proximities among domains and IPs. Extensive experiments on real DNS traffic demonstrate that the joint optimization of multiple tasks with the latent high-order proximities can lead to better security analysis performance for all the tasks than respectively optimizing each single task with the observable low-order proximity.
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.