Learning and reasoning method of hybrid Bayesian network

Abstract

The invention relates to a learning and reasoning method for a hybrid Bayesian network, and the method comprises the steps: S1, the discretization of a continuous variable is carried out through employing a supervised top-down discretization algorithm CACC; s2, a Gibbs Sampling algorithm is adopted to solve the parameter learning problem of an incomplete data set, and a complete hybrid Bayesian network model is obtained; and S3, a Markov chain Monte Carlo algorithm (MCMC) and a joint tree algorithm are adopted as a hybrid Bayesian network inference algorithm, and an inference algorithm more suitable for an application scene is sought. The continuous variables are discretized by adopting the CACC algorithm, so that the calculation efficiency is ensured; according to the parameter learning method of the incomplete data set adopted by the invention, the parameter learning steps are simplified, the network construction period is shortened, and the occurrence of extreme data is avoided. Thetwo hybrid Bayesian network reasoning methods adopted by the invention are good in accuracy and timeliness.

Description

technical field [0001] The invention relates to the field of abnormal detection and positioning of Bayesian networks, and more specifically, to a learning and reasoning method of mixed Bayesian networks. Background technique [0002] Bayesian network is one of the most effective theories in the field of uncertain knowledge representation and reasoning, and is widely accepted because of its friendly explanation of causality. Its main idea is to take into consideration the process rules during the production or operation of mechanical equipment, use statistical thinking to process historical data, transform influencing factors into dependencies, and fully integrate theoretical knowledge with knowledge of practical application scenarios. For Bayesian networks that contain both discrete nodes and continuous nodes, they are generally referred to as Hybrid Bayesian Networks. Due to the particularity of its continuous nodes, the hybrid Bayesian network generally discretizes the da...

AI Technical Summary

Problems solved by technology

[0007] The present invention provides a hybrid Bayesian network learning and reasoning method to overcome the shortcomings of the Bayesian network-based learning and reasoning methods described in the prior art, which are not accurate enough and timely.

Images