摘要
根据中医诊断推理对知识的不确定性、不完全性和逻辑推理的模糊性及“软计算”技术的特点出发,探讨了利用“软计算”的方法构建中医诊断神经网络知识库的基本设想,最后在MATLAB中利用现有结构固定的三种神经网络模型分别粗略地构造了中医“八纲辨证”神经网络,并分析其计算结果。说明采用神经网络技术构建中医诊断知识库是完全可行的。
Traditional Chinese-Medical diagnosis has the characteristic of uncertainty, imperfection and illegibility, and the soft compute has the same characteristic. Form the point of view, we discuss the thought that use the technology of soft compute to construct the Traditional Chinese-Medical NN( Neural Network) knowledge base. In the thesis, we also use three models of neural network, which have the fixed frame, to construct the neural network of the TCM tongue diagnosis eight principal syndromes, and analyses the result. So it' s feasible that using the NN technology to construct the TCMD knowledge base.
出处
《计算机应用研究》
CSCD
北大核心
2006年第6期188-189,198,共3页
Application Research of Computers
基金
国家自然科学基金资助项目(60275023)
厦门大学院士基金资助项目
关键词
软计算
神经网络
中医诊断
中医“八纲辨证”知识库
Soft Compute
NN
TCMD(Traditional Chinese-Medical Diagnosis)
TCM Eight Principal Syndromes NN Knowledge Base