摘要
采用模糊逻辑系统和产生式神经网络作为工程型知识库系统中信息平行处理的基础。在模糊决策系统中,状态空间中每个决策步骤依据知识库中的规则集,按模糊匹配度予以评价和平行处理。定义网络结点物理意义的产生式神经网络可改善其处理效果。由选择分离器和化学反应器两个实例阐明了平行处理的原理和方法。
In this paper,a fuzzy logic system and a productive neural network are used as the basis of parallel information processing and inference in knowledge base systems.In the fuzzy decision_making systems,decisions are evaluated by the fuzzy set containing match,which is an index of how it conforms with the rule set in the knowledge base.In another approach,parallel information processing in the neural network was improved by assigning physical meanings to network nodes.Two examples in the selection of chemical...
出处
《华南理工大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
1997年第7期9-14,共6页
Journal of South China University of Technology(Natural Science Edition)
基金
国家自然科学基金
国家教委优秀年轻教师基金
关键词
知识库
推理
信息处理
模糊匹配
神经网络
knowledge base
inference
information processing
fuzzy match
neural network