Evolving fuzzy linear regression trees with feature selection

Abstract

This paper introduces an approach to evolve fuzzy modeling that simultaneously performs adaptive feature selection. The model is a fuzzy linear regression tree whose topology can be continuously updated using statistical tests. A fuzzy linear regression tree is a fuzzy tree with linear model in each leaf. The number of tree nodes and the number of inputs can be updated for each new input. The precision and the feature selection mechanism of the proposed model are evaluated using system identification and time series forecasting problems. The results suggest that the evolving tree model is a promising approach for adaptive system modeling with feature selection.