Feature Confidence Guiding Incremental Learning Algorithm of Bayesian Network
Songhao Jia, Cai GuoqiangJianwei YANG - · International Journal of Applied Mathematics & Statistics/International journal of applied mathematics and statistics · 2013
Current Bayesian Network incremental learning methods can not be dynamic tracking of user needs, and its performance is not high. As a result, reliability of the Bayesian network can not be guaranteed. In order to solve the problem, this paper introduces the guarding of feature confidence to incremental learning, and puts forward the feature confidence guiding incremental learning algorithm of Bayesian Network (FCG-IBN). The algorithm is proposed to improve the learning method of Bayesian Network, enhance each batch data the precision of the study, and improve the quality of the final network model. The result of the experiments indicates that the combination of FCG-IBN can learn better accurate network from datasets. The application results have proved that it is worth promoting the use.