基于粗糙集和神经网络结合的鱼病诊断方法

Fish diseases diagnosis based on rough set and neural network

为了实现鱼病的快速和大批量诊断的目的,将粗糙集理论和神经网络紧密结合建立一种新的高效鱼病诊断模型。利用粗糙集进行知识获取,即把鱼病的典型案例作为样本空间形成"症状—疾病"的决策表,然后根据粗糙集的知识简化方法,去掉冗余的属性和样本。利用性能优良的模糊kohonen聚类网络进行聚类分析,最后形成鱼病的分类规则,新的鱼病就可通过此规则进行诊断。该模型充分融合了粗糙集强大的规则提取能力和神经网络优良的分类能力,实验证明模型具有很好的分类效率,可以实现鱼病的快速诊断。

In order to achieve the rapid and mass diagnosis of fish diseases, a new high-performance model is set up, which closely connects rough set and neural network. First, the rough set is used for access to knowledge, that is, the typical cases of fish diseases are regarded as sample room for the formation of the decision-making table of the ＂symptoms--disease＂. Next, based on rough set of simplified method of knowledge, redundant properties and samples are removed. Then, the fine performance of the fuzzy kohonen clustering network is used to analyze clustering; and finally fish diseases classification rules are formed. The model integrate the strong extracting capabilities of rough set and the excellent classifying ability of neural network, and is proved experimentally to be efficient in classification and rapid in fish diseases diagnosis.