摘要
在森林培育专家系统中,由于一些知识的不完整性和模糊性,造成模拟专家推理过程中推理结果的不确定。为实现对这些知识的处理,将不确定性推理应用于森林培育专家系统,利用模糊逻辑方法对确定性和不确定性知识用多个逻辑值进行表达,并在规则可信度的计算中,将专家给出的特定规则的可信度规定为1;将专家给出的不确定规则,通过频率来动态计算获得规则的权重值和可信度;而对于用户不确定的选择而产生的规则,采用模糊层次分析法计算可信度。另外,本研究的推理机算法实现了根据用户上一条的回答智能判断下一条提问,并基于用户的回答,给出该立地条件下适宜的造林树种或造林模式以及可信度。
In the forest cultivation expert system, reasoning results are usually made in the face of uncertainty, with knowledge that is incomplete or fuzzy. Uncertainty inference was applied in the forest cultivation expert system to deal with such knowledge. Fuzzy logic was used to express different degrees of certainty or uncertainty of assertions. The certainty factor of specific rules by experts given equalled 1 ; the weight factor and certainty factor of the uncertainty rule were calculated from the frequency; the certainty factor of rules was calculated using the fuzzy-analytic hierarchy process for the users＇ uncertain selection. Furthermore, an inference engine algorithm was designed. The order of questions to be asked was decided dynamically depending upon the answers of the users. On the basis of answers, consulting resuhs along with its confidence factor were provided.
出处
《北京林业大学学报》
CAS
CSCD
北大核心
2014年第1期88-93,共6页
Journal of Beijing Forestry University
基金
"863"国家高技术研究发展计划项目(2012AA102003)
国家自然科学基金项目(31170513)
关键词
可信度
推理机算法
专家系统
不确定性推理
certainty factor
inference engine algorithm
expert system
uncertainty inference
