TIME-SERIES ANALYSIS BY KAUFFMAN NETWORKS

This paper describes a way by which a Kauffman Network can be identifi ed so that it can be used to model a system. The only known characteri stics of the system is a time series of observations over the system. Fourier analysis is applied to the time series to determine the size o f the Network, the values that the cells may assume, and the transitio n rules. Results of several experiments are reported and found to be s upportive to the method derived.