A new PC-PSO algorithm for Bayesian network structure learning with structure priors

作者：

Highlights：

• A new Bayesian network structure learning algorithm has been proposed.

• This method is based on the hybrid approach, which combines two classic algorithms.

• Two operators of Genetic Algorithm are used to search feasible solutions.

• This method outperforms conventional algorithms in the experiments.

摘要

•A new Bayesian network structure learning algorithm has been proposed.•This method is based on the hybrid approach, which combines two classic algorithms.•Two operators of Genetic Algorithm are used to search feasible solutions.•This method outperforms conventional algorithms in the experiments.

论文关键词：Bayesian networks,Structure learning,Particle swarm optimization,PC algorithm,Structure priors

论文评审过程：Received 9 May 2020, Revised 10 September 2020, Accepted 16 May 2021, Available online 11 June 2021, Version of Record 28 June 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115237