陶瓷制品缺陷神经网络型知识库的研究

On Neural Network Knowledge Base of Fault Analysis of Ceramic Products

探讨由陶瓷制品缺陷分析样本来建造神经网络,它可作为专家系统的知识库。对专家系统的知识获取问题,神经网络给出了一条新的途径。它适用于分类、诊断和鉴别问题,且适用于诸如过程控制的实时应用环境。

In this paper we have examined how to construct a neural network from training samples and how to use it as the knowledge base for an expert system. It presents a promising approach to the knowledge-acquisition problem for expert system. These methods are most appropriate for diagnosis and classification problems. This makes neural networks expert system especially well suited for real-time applications such as process control.