ANFIS Based Expert Systems Prototype Development for Estimation of Urban Mobility

Abstract

For the purpose of estimating urban mobility, an approach that uses urban mobility indicators calculated from the data subsystem from the Charging Data Records database in the public mobile telecommunications network was chosen. Calculated indicators are unified in the urban mobility index as a unique indicator of urban mobility in urban agglomeration. The urban mobility index is calculated using an expert system based on fuzzy logic, that is, the adaptive neuro-fuzzy inference system method (ANFIS). It functions on the principle of applying conclusion methods which characterize neural networks with the goal of determining parameters of an indirect conclusion system (Fuzzy Inference System - FIS). The prototype of the system was developed by using the appropriate application environment, where the proposed methodology was implemented.