An application of online learning algorithm for Bayesian network parameter
Shaozhong Zhang, Hong Qing Yu, Hua Feng Ding, Nanhai Yang, Xiu-Kun Wang · 2004
Bayesian network is a graphical model that encodes probabilistic relationships among nodes of interest. The automated creation of Bayesian networks can be separated into two tasks. Structure learning, which consists of creating the structure of the Bayesian networks from the collected data and parameter learning, which consists of calculating the numerical parameters for a given structure. A Voting EM algorithm which is based EM be discussed and applied in the online parameter learning in flood decision Bayesian networks in this paper. Both EM and Voting EM algorithm are applied in flood decision Bayesian networks to compared their performance. The result indicates that the Voting EM can be used in the online learning for Bayesian network parameter and it also has more precisely that general EM algorithm.