The Common Solution And Qualities Of Uncorrelated Linear Discriminant Analysis Based On Singular Value Decomposition

High-dimensional data always appear in the fields of Data mining,Machine learning and Bioinformatics and so on.In order to overcome the problem of high dimension,the conventional method we take is reducing the dimension of high-dimensional data.Currently,uncorrelated linear discriminant analysis is a good way to reduct dimension.Because of the uncorrelated quality of the feature vectors,this approach do best in reducing the data redundancy. In this paper,firstly,we use two methods to solve the problem of uncorrelated discriminant analysis,that is the way based on the Singular value decomposition and the way based on QR decomposition.And in the process of calculation,we get a new conclusion suprisingly.This passage is organised as follows: In the first part we briefly review LDA and U LDA;Then,in the second and third part we give singular value decomposition algorithm and QR decomposition algorithm;Finally,we explain the relationship between the two algorithms in part 4.