| 一种新的快速模糊规则提取方法 |
| |
| 引用本文： | 陈铁明,龚荣盛,SAMUEL H Huang.一种新的快速模糊规则提取方法[J].控制与决策,2008,23(9). |
| |
| 作者姓名： | 陈铁明 龚荣盛 SAMUEL H Huang |
| |
| 作者单位： | 1. 浙江工业大学软件学院,杭州,310014;北京航空航天大学软件开发环境国家重点实验室,北京100086
2. 辛辛那提大学工程学院,美国,辛辛那提,45221 |
| |
| 基金项目： | 国家自然科学基金，国家高技术研究发展计划(863计划)，浙江省自然科学基金 |
| |
| 摘 要： | 提出一种高效的规则提取算法,采用熵测量改进Chi-merge特征区间离散化方法,模糊划分输入空间闻.先为每个数据生成单条规则,再聚集相同前项的单条规则产生带概率属性的分类规则.提取的规则无需任何调整,应用模糊推理便可获得较理想的分类效果,同时支持增量规则更新.最后给出了新方法的性能测试结果.
|
| 关 键 词： | 模糊规则 规则提取 Chi-merge方法 离散化 模糊推理 |
A novel and quick fuzzy rule extraction method |
| |
| CHEN Tie-ming,GONG Rong-sheng,SAMUEL H Huang.A novel and quick fuzzy rule extraction method[J].Control and Decision,2008,23(9). |
| |
| Authors: | CHEN Tie-ming GONG Rong-sheng SAMUEL H Huang |
| |
| Affiliation: | CHEN Tie-ming1,2,GONG Rong-sheng3,SAMUEL H Huang3(1.College of Software Engineering,Zhejiang University of Technology,Hangzhou 310014,China,2.State Key Laboratory of Software Development Environment,Beihang Univerity,Beijing 100086,3.School of Engineering,University of Cincinnati,Cincinnati 45221,USA.) |
| |
| Abstract: | This paper presents a novel method for classification problems.Firstly,entropy is introduced to select the best discretization solution from Chi-merge for each feature to guide grid partition.Afterwards one simple rule is generated from each data point,and the combination of simple rules with the same antecedent yields the consequent of an aggregated rule for each grid.Because grid partition is based on rational discretization,the extracted rules are not only easy to understand but also produce accurate res... |
| |
| Keywords: | Fuzzy rules Rule extraction Chi-merge method Discretization Fuzzy inference |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
