Internet Data for GNN Benchmarking
收藏arXiv2022-10-26 更新2024-06-21 收录
下载链接：
https://github.com/dpgiakatos/gnn-internet-data
下载链接
链接失效反馈官方服务：
资源简介：
本数据集由亚里士多德大学编译，旨在为互联网数据上的图机器学习和图神经网络研究提供基准。数据集包含19种异构节点/AS属性，来源于多个在线资源，并经过预处理以适应GNN架构的输入。数据集不仅便于研究者访问，还作为基准数据集，支持不同机器学习方法的应用和比较，特别适用于解决互联网路由和操作相关问题。
This dataset was compiled by Aristotle University to provide benchmarks for graph machine learning and graph neural network research on Internet data. It contains 19 types of heterogeneous node/AS attributes sourced from multiple online resources, and has been preprocessed to fit the input requirements of GNN architectures. The dataset is not only easily accessible to researchers, but also serves as a benchmark dataset that supports the application and comparison of various machine learning methods, and is particularly suitable for addressing issues related to Internet routing and operations.
提供机构：
亚里士多德大学
创建时间：
2022-10-26
