A Knowledge Discovery Method Based on Error Matrix Equation

This paper begins by defining error matrix to model system's interacting objects whose microscopic state includes not only spatio-temporal variables but also error functions. The error matrix model allows us to define six transformations that have been proposed by error-eliminating theory's preliminary researches. The main result of this paper is a set of error matrix equations such as T(u) = u 1 . The relative solution is given herein. There are ten equations are defined in this paper. These equations are divided into 2 types and there are 5 kinds of operators in each type. Error matrix is used to express current status u, expectant status u 1 and transformation T. It is u, u 1 , and T that are used to build error matrix equation. The research results provide a new useful potential technique for the analysis of social problems. It allows us to find the method that bad status ¿ u¿ change to good status ¿u 1 ¿ by means of the solution T.