Applications of bio-inspired systems as function approximators
Bogdan Fîrtat, S. Lazar · 2004
In this paper we would like to show how the bio-inspired systems topologies can be applied as function approximators (nonlinear regressors) in a wealth of practical applications. We selected one application in the financial industry and another in real estate. The goal is to discover nonlinear mappings between input variables and important outcomes. In the financial arena the outcome is to predict the value of the S&P 500 (one of the most important Standard&Poor's indexes), using several financial indicators, while in the real estate application the problem is to estimate the price of a house based on several indicators. We will see that bio-inspired systems provide a very powerful analysis tool. The software tool used was NeuroSolutions, from NeuroDimension Inc., a leading edge neural network development software.