| 基于神经元网络的不确定性规则的获取 |
| |
| 引用本文： | 王红卫,费奇. 基于神经元网络的不确定性规则的获取[J]. 华中科技大学学报(自然科学版), 1993, 0(2) |
| |
| 作者姓名： | 王红卫 费奇 |
| |
| 作者单位： | 华中理工大学系统工程研究所(王红卫)，华中理工大学系统工程研究所(费奇) |
| |
| 基金项目： | 国家自然科学基金资助项目 |
| |
| 摘 要： | 讨论了不确定性规则的知识系统同神经元网络相结合的问题;提出了适合处理不确定性规则的负梯度学习算法,使不确定性规则能方便地转换成神经元网络的权矩阵;研究了如何利用神经元网络的生成能力来获取新的不确定性规则,使知识系统具有知识自动获取能力,并使原有规则系统的脆弱性得以改善.
|
| 关 键 词： | 神经元网络 不确定性规则 负梯度学习算法 可信度 知识系统 |
Acquisition of Uncertainty Rules Based on Neural Networks |
| |
| Wang Hongwei Fei Qi. Acquisition of Uncertainty Rules Based on Neural Networks[J]. JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE, 1993, 0(2) |
| |
| Authors: | Wang Hongwei Fei Qi |
| |
| Affiliation: | Wang Hongwei Fei Qi |
| |
| Abstract: | This paper is concerned with the combination of the knowlede system of uncertainty rules with neural networks. A negative gradient learning algorithm suitable for handling uncertainty rules is proposed. This makes it convenient to transform the uncertainty rules into a weight matrix of the neural network. A method of acquiring new uncertainty rules by the generating ability of the neural network is also discussed. Hence the knowledge system is enabled to acquire knowledge automatically and the rule system shortcoming,the vulnerability, can be removed. |
| |
| Keywords: | neural network uncertainty rule negative gradient learning algorithm -degree of confidence knowledge system |
| 本文献已被 CNKI 等数据库收录！ |
