A novel framework of graph Bayesian optimization and its applications to real-world network analysis

作者：

Highlights：

• A novel framework GBO for the graph structure optimization has been proposed.

• Both structural information and global information on graphs are explored by GBO.

• A novel problem of opening the gated residential areas is presented.

• Several instructive rules are discovered for expert decision-making.

• GBO can be applied to a variety of practical network problems empirically.

摘要

•A novel framework GBO for the graph structure optimization has been proposed.•Both structural information and global information on graphs are explored by GBO.•A novel problem of opening the gated residential areas is presented.•Several instructive rules are discovered for expert decision-making.•GBO can be applied to a variety of practical network problems empirically.

论文关键词：Graphs,Networks,Structure optimization,Bayesian optimization,Graph kernels

论文评审过程：Received 2 March 2019, Revised 6 June 2020, Accepted 18 December 2020, Available online 25 December 2020, Version of Record 18 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114524