Neuro-Fuzzy Projection Pursuit Regression

Abstract

The projection pursuit (PP) for regression was originally proposed by Friedman and Stuetzle. The important feature of PP is that it is one of the multivariate methods able to bypass the "curse of dimensionality". The aim of PP is to find an interesting or characteristic structure by working in low-dimensional linear projections. In this paper, we propose a method based on the membership function and the eigenvector of the covariance matrix to avoid the local minimum of the projection indices.