CVE Entries from NVD

该数据集包含了5,637个来自国家漏洞数据库（NVD）的CVE条目，旨在训练一个多标签分类模型，以预测网络安全漏洞的严重性和类型。数据集包括了将严重性标签（低、中、高、关键）转换为数值索引，以及将10个预定义类别中的类型编码为多热向量。此外，该数据集通过分层抽样方法被划分为80%的训练集和20%的验证集。规模上，数据集包含了5,637个CVE条目，任务是对常见漏洞和暴露（CVE）报告进行多标签分类。

This dataset contains 5,637 CVE entries sourced from the National Vulnerability Database (NVD), and is designed to train a multi-label classification model for predicting the severity and type of cybersecurity vulnerabilities. The dataset includes converting severity labels (low, medium, high, critical) into numerical indices, as well as encoding vulnerability types from 10 predefined categories into multi-hot vectors. Additionally, this dataset is split into an 80% training set and a 20% validation set via stratified sampling. In terms of scale, the dataset encompasses 5,637 CVE entries, and the targeted task is to perform multi-label classification on Common Vulnerabilities and Exposures (CVE) reports.