Pattern Recognition for Ship Based on Bayesian Networks
Wang Qingjiang, Xiaoguang Gao, DaQing Chen · 2007
Bayesian networks (BNs) are a powerful tool for pattern recognition. A BNs has two parts: parameters and structure composed of a directed acyclic graph (DAG) with some nodes. Then, according to the target feature, an approach based on BNs for pattern recognition is presented and the step of the approach is presented: constructing its nodes, modifying the node's states and distributing the node's probability. The process using the approach for pattern recognition is showed by an experiment, and the empirical results provide evidences that the approach is reasonable and effective.