属性值为区间灰数的判别分析法  

Analysis Test when the Data is Grey Number

作　　者：王志国[1] 李冬峰[1] 

机构地区：[1]华北水利水电学院,河南郑州450011

出　　处：《中国西部科技（学术版）》2007年第3期9-10,共2页Science and Technology of West China

摘　　要：判别分析法是在已知研究对象分成若干类型并已取得各种类型的一批已知样品的观测数据,在此基础上根据某些准则建立判别式,然后对未知类型的样品进行判别分类。当观测数据是实数时,我们可以很容易地按照已有的判别方法进行归类,但在实际情况中往往不容易测得精确数据,而是一个个区间灰数。在这种情况下,本文采取灰色系统理论中将灰数白化的方法,先将灰数白化,然后再按照判别分析法进行判别归类。The analysis test is such a method that divides the unrehearsed objects into several types whose sample data have been achieved and establishes test according to some criteria, then tests and classifies the sample of unknown type, We can classify them easily according to previous tests when the data are real numbers. It is not easy to get accurate data in realistic situation but grey data, Under this situation, we can white the grey data at the beginning,then classify them according to ordinary analysis test.

关 键 词：区间灰数 判别分析法 白化 

分 类 号：N94[自然科学总论—系统科学]