Comparison of Adaptive Network Based Fuzzy Inference Systems and B-spline Neuro-Fuzzy Mode Choice Models

Abstract

This paper investigates the use of neuro-fuzzy models for behavioral mode choice modeling. The concept of neuro-fuzzy models has emerged in recent years as researchers have tried to combine the transparent, linguistic representation of a fuzzy system with the learning ability of artificial neural networks. Several neuro-fuzzy systems have been reported in the literature. They include various representations and architectures and therefore are suitable for different applications. In this paper, the performance of two of the most widely used neuro-fuzzy models, namely: B-spline associative memory networks and adaptive network based fuzzy inference systems, is compared. The theoretical backgrounds of both systems are presented and their relative advantages are discussed using a mode choice modeling case study. Areas of comparison include: model performance, dealing with the curse of dimensionality, automatic exclusion of irrelevant inputs, and model transparency.