FUZZY REGRESSION BITS Pilani Hyderabad Campus TEAM MEMBERS • • • • ANUPAMA HARIKA TRIKUTAM TARUN GHANTA REVANTH GADAMSETTY BITS Pilani, Hyderabad Campus INTRODUCTION The Regression analysis is a technique of studying the dependence of one variable, on one or more variables, with a view to estimate or predict the average value of the dependent variables in terms of the known or fixed values of the independent variables. The goal of the regression analysis is: (a) to find an appropriate mathematical model, and (b) to determine the best-fitting coefficients of the model from the given data. BITS Pilani, Hyderabad Campus Cont….... The following model shows the dependence of the output variable on the inputs variables: (1) BITS Pilani, Hyderabad Campus Cont….... BITS Pilani, Hyderabad Campus Cont….... BITS Pilani, Hyderabad Campus NON-SYMMETRIC FUZZY REGRESSION BITS Pilani, Hyderabad Campus FULLY FUZZY LINEAR REGRESSION SYSTEM BITS Pilani, Hyderabad Campus MULTIPLICATION OF TWO FUZZY NUMBERS BITS Pilani, Hyderabad Campus METHODOLOGY Methods used to solve the fully fuzzy linear regression ❑ Huang’s method ❑ LU decomposition method ❑ QR decomposition BITS Pilani, Hyderabad Campus HUANG’S METHOD . . . . . . BITS Pilani, Hyderabad Campus Cont….... (A,M,N)⊗(x,y,z) = (b,g,h) (Ax, Ay+Mx,Az+Nx)=(b,g,h) BITS Pilani, Hyderabad Campus Cont….... Ax=b Ay+Mx=g Az+Nx=h BITS Pilani, Hyderabad Campus Cont….... HUANG’S METHOD METHOD ALGORITHM Step 1: Express fuzzy matrix as three crisp matrices A,M,N • Take Step 2 : Substitute A,M,N in equation • (A,M,N)⊗(x,y,z) = (b,g,h) BITS Pilani, Hyderabad Campus Cont….... Step 3 : compute x,y,z from the following equations •Ax=b ⟹ x= •Ay=g-Mx ⟹•Az=h-Nx ⟹ z=- BITS Pilani, Hyderabad Campus Cont….... (code and examples of Huang’s method to be continued on Python) BITS Pilani, Hyderabad Campus LU DECOMPOSITION METHOD A= LU BITS Pilani, Hyderabad Campus Cont….... BITS Pilani, Hyderabad Campus LU DECOMPOSITION METHOD ALGORITHM Step 1: Assume that = (A,M,N) A= Step 2: Set Step 3: Compute Step 4: Compute BITS Pilani, Hyderabad Campus Cont….... Step 5: Compute the solution of the fully fuzzy linear system (A,M,N)⊗(x,y,z)=(b,g,h) as follows: x= y= z= BITS Pilani, Hyderabad Campus Cont….... (code and examples of LU decomposition method to be continued on Python) BITS Pilani, Hyderabad Campus QR DECOMPOSITION THEOREM: If A is an m×k matrix with full column rank, then A can be factored as A = QR where Q is an m×k matrix whose column vectors form an orthonormal basis for the column space of A and R is a k×k invertible upper triangular matrix. BITS Pilani, Hyderabad Campus Cont….... BITS Pilani, Hyderabad Campus Cont….... (A,M,N)⊗(x,y,z) = (b,g,h) BITS Pilani, Hyderabad Campus APPLICATION OF FUZZY REGRESSION Estimation of air transport demand Real state price prediction Prediction of tool life used in milling BITS Pilani, Hyderabad Campus Cont….... Predict exchange rates for composite currencies Prediction of dissolved oxygen Market demand estimation for new product development BITS Pilani, Hyderabad Campus THANK YOU BITS Pilani, Hyderabad Campus