基于精简模糊分类关联规则的分组模糊判决方法  

Grouping fuzzy decision based on simplified fuzzy classification association rules

作　　者：杨敬锋[1] 薛月菊[1] 胡月明[2] 陈志民[3] 陈强[1] 

机构地区：[1]华南农业大学工程学院,广州510642 [2]华南农业大学信息学院,广州510642 [3]华南农业大学公共基础课实验教学中心,广州510642

出　　处：《系统工程理论与实践》2008年第3期139-143,共5页Systems Engineering-Theory & Practice

基　　金：广东省自然科学基金(04300504,980150);国家科技攻关项目(2002BA516A08);广东省科技攻关项目(2005B20701008,2005B10101028,2004B20701006);华南农业大学校长基金(2007K017)

摘　　要：为提高土地评价知识表达的简易性和可解释性,提出利用精简模糊分类关联规则和模糊判决进行土地评价的方法.为了降低土地评价模型的复杂程度,提高模糊关联规则分类的有效性和可解释性,本文通过精简模糊分类关联规则,去除了冗余规则,并针对了模糊判决中难以判决的问题,提出分组模糊判决算法进行迭代.实验表明,在采用32条精简规则的情况下,结合精简模糊分类关联规则和分组模糊判决进行土地评价方法获得准确率为92.2835%,比精简前在最小支持度为0.005的情况下得到的32条模糊分类关联规则准确率提高了5.0039%.To improve the intelligibility and efficiency of knowledge expression for the land evaluation, a land evaluation method combining simplified fuzzy classification association rules with fuzzy decision is proposed in this paper. To reduce the complexity of the land evaluation models and improve the efficiency and intelligibility of fuzzy classification association rules further, an algorithm to eliminate redundant rules for obtaining the simplified fuzzy classification association rules is presented. In addition, considering the challenge of a few samples that are difficult to classify the process of fuzzy decision, an iterative algorithm for grouping fuzzy decision for datasets is discussed. The results of experiments demonstrate that by using only 32 simplified fuzzy classification association rules, accuracy of area of land evaluation can reach 92.2835 percent. It provides a higher precision with the accuracy improved by 5. 0039 %, comparing with the results of the method combining 32 original fuzzy classification association rules with fuzzy decision when minimum support is 0.005.

关 键 词：模糊分类关联规则 精简规则 模糊判决 分组模糊判决 土地评价 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]