The VNF cybersecurity dataset for research (VNFCYBERDATA)
收藏Figshare2024-11-07 更新2026-04-28 收录
下载链接:
https://figshare.com/articles/dataset/The_VNF_cybersecurity_dataset_for_research_VNFCYBERDATA_/24998543
下载链接
链接失效反馈官方服务:
资源简介:
Virtualisation has received widespread adoption and deployment across a wide range of enterprises and industries throughout the years. Network Function Virtualisation (NFV) is a technical concept that presents a method for dynamically delivering virtualised network functions as virtualised or software components. Virtualised Network Function (VNF) has distinct advantages, but it also faces serious security challenges. Cyberattacks such as Denial of Service (DoS), malware/rootkit injection, port scanning, and so on can target VNF appliances just like any other network infrastructure. To create exceptional training exercises for machine or deep learning (ML/DL) models to combat cyberattacks in VNF, a suitable dataset (VNFCYBERDATA) exhibiting an actual reflection, or one that is reasonably close to an actual reflection, of the problem that the ML/DL model could address is required. This article describes a real VNF dataset that contains over seven million data points and twenty-five cyberattacks generated from five VNF appliances. To facilitate a realistic examination of VNF traffic, the dataset includes both benign and malicious traffic.CitationIf you are using this dataset for your research, please reference it as"Ayodele, B.; Buttigieg, V. The VNF Cybersecurity Dataset for Research (VNFCYBERDATA). Data 2024, 9, 132. https://doi.org/10.3390/data9110132"DocumentationDataset documentation is available at: https://www.mdpi.com/2306-5729/9/11/132
多年来,虚拟化技术已在全球各类企业与行业中得到广泛采用与部署。网络功能虚拟化(Network Function Virtualisation, NFV)作为一种技术理念,提供了将虚拟化网络功能以虚拟化或软件组件形式动态交付的方案。虚拟化网络功能(Virtualised Network Function, VNF)虽具备显著优势,但同时也面临严峻的安全挑战。拒绝服务(Denial of Service, DoS)、恶意软件/rootkit注入、端口扫描等网络攻击,均可如同针对其他网络基础设施一般,靶向攻击VNF设备。
为构建可用于训练机器学习/深度学习(Machine/Deep Learning, ML/DL)模型以对抗VNF环境下网络攻击的优质训练数据集,需要一个能够真实反映(或尽可能贴近真实场景)该ML/DL模型所要解决问题的合适数据集(VNFCYBERDATA)。本文介绍了一个真实的VNF数据集,该数据集包含超过700万个数据点,涵盖由5台VNF设备生成的25种网络攻击场景。为便于对VNF流量开展真实场景下的分析,该数据集同时包含良性流量与恶意流量。
引用说明:若您在研究中使用该数据集,请按以下格式引用:Ayodele, B.; Buttigieg, V. 面向研究的VNF网络安全数据集(VNFCYBERDATA). Data 2024, 9, 132. https://doi.org/10.3390/data9110132
数据集文档:数据集文档可通过以下地址获取:https://www.mdpi.com/2306-5729/9/11/132
创建时间:
2024-11-07



