A Prufer-leaf Coding Genetic Algorithm For Bayesian Network Structure Learning
Ying Yu, Shili Luo, Yanru He, Hao Huang, Wei Zhang · 2022 Global Conference on Robotics, Artificial Intelligence and Information Technology (GCRAIT) · 2022
All Bayesian network is a probabilistic graph model which is used to describe the causal distribution among variables and has been widely used in many fields. While Bayesian network structure learning is the core of Bayesian network research, it is also an NP-hard problem. This paper draws on the idea of Prufer encoding/decoding, combines genetic algorithm and mutual information theory, and proposes a genetic algorithm based on improved Prufer-Leaf coding to solve the learning optimization problem of Bayesian network structures. Finally, the superiority of the algorithm in learning Bayesian network structure is verified through experiments.