新的Hopfield神经网络分类器在葡萄酒质量评价中的应用

摘要

建立一种新的Hopfield神经网络分类器模型,该模型通过训练单层前向神经网络来设计,数据兼容性强,可以直接处理来自UCI数据库的葡萄酒的理化性质测试指标数据和专家的感官评价等级数据,实现葡萄酒质量分类.仿真结果表明,该分类器设计简单,耗时短,分类效果明显.%A model of the quality evaluation of grape wine based on Hopfield neural network classifier is proposed in this paper. The design of model is achieved by training a single-layer feedforward network. The model is compatible with data from UCI dataset. The data are composed of physicochemical properties and sensory test results of wine. The simulation results show that the design of classifier model is simple and it achieves effective class outcome.