CIC IDS-2017

该数据集是一个网络入侵数据集，包含了良性流量和常见的现实世界攻击类型。它由PCAP格式的原始流量数据文件和CSV格式的流量特征及分类标签文件组成。数据集涵盖了七大类攻击：基于Web的攻击、暴力破解、拒绝服务（DoS）、分布式拒绝服务（DDoS）、渗透、Heartbleed漏洞利用和僵尸网络。此外，它还融入了基于专家对入侵检测相关流量特性的知识所工程化的特征。该数据集规模宏大，包含了多天的网络流量数据（相当于五天的PCAP文件）。其任务是网络入侵检测。

This is a network intrusion dataset containing benign network traffic and common real-world attack types. It consists of raw traffic data files in PCAP format and CSV-format files that store traffic features and classification labels. The dataset covers seven major categories of attacks: web-based attacks, brute-force attacks, Denial of Service (DoS), Distributed Denial of Service (DDoS), penetration attacks, Heartbleed exploit, and botnets. Furthermore, it incorporates features engineered based on expert knowledge of traffic characteristics relevant to intrusion detection. This large-scale dataset contains multiple days of network traffic data, equivalent to five days' worth of PCAP files. The task associated with this dataset is network intrusion detection.

搜集汇总

背景与挑战

背景概述

CIC IDS-2017是一个用于网络安全威胁检测的数据集，包含带标签的数据包，旨在支持基于原始数据的机器学习模型训练。该数据集关联于一篇研究人工智能方法在计算机网络中网络安全威胁检测的论文。