摘要
论文简要介绍了粗糙集依赖度的概念,认为用依赖度作为粗糙集敏感度分析不完全正确,而用粗糙集不协调率进行敏感度分析能够准确地评价各因素的影响,并构造了粗糙集化简树。根据粗糙集化简树能够准确地找到所有的最终条件属性简化集,并计算出各因素的粗糙集不协调率。根据粗糙集不协调率可以精确地判断实际工程应用中诸多的影响因素的重要程度,通过粗糙集化简树为分析、简化复杂的知识信息系统提供了一条新的途径,并为决策者作出决策提供了可靠依据。
This paper simply introduces the conception of Rough Sets Relevancy Degree,and points out that it is not correct to use Rough Sets Relevancy Degree to value the significance of factors,but to use Non-correspond Ratio of Rough Sets replacing it can get good results.By the way,the predigesting tree is created.It can well and truly gain all the last predigesting tables,and easily count the Non-correspond Ratio of Rough Sets.To judge the factors of effecting which base on the non-correspond ratio of rough sets to find the main factors,and the predigesting tree of Rough Sets which represent and simplify knowledge as a tool in Decision Support System(DSS).
出处
《计算机工程与应用》
CSCD
北大核心
2003年第18期49-50,198,共3页
Computer Engineering and Applications
基金
国家自然科学基金资助(编号:10071010)
关键词
依赖度
粗糙集不协调率
决策表
简化树
敏感度
Relevancy Degree,Non-correspond-ratio of Rough Sets,Decision Table,Predigesting tree,Sensitivity Degree
