Inference in hybrid Bayesian networks with large discrete and continuous domains
作者：
Highlights：
• Bayesian inference algorithm for a large domain of discrete variables is proposed.
• Classification tree algorithm identifies conditional probability tables.
• Tree based factor production and marginalization methods are proposed.
• Efficient inference algorithm for tree structured Bayesian network is developed.
• Estimation of most likely production times for steel plates production.
摘要
•Bayesian inference algorithm for a large domain of discrete variables is proposed.•Classification tree algorithm identifies conditional probability tables.•Tree based factor production and marginalization methods are proposed.•Efficient inference algorithm for tree structured Bayesian network is developed.•Estimation of most likely production times for steel plates production.
论文关键词：Large domain of discrete and continuous variables,Decision tree,Hybrid Bayesian network,Belief propagation algorithm,Context-specific independence
论文评审过程：Received 15 July 2015, Revised 19 November 2015, Accepted 20 November 2015, Available online 2 December 2015, Version of Record 24 December 2015.
论文官网地址：https://doi.org/10.1016/j.eswa.2015.11.019
