Detection of dynamical structure from short and noisy chaotic series

A method is proposed to detect the dynamical structure hiding behind complex chaotic series by comparing prediction performance of trial functions. This method is valid even when the original system is contaminated with noise or only a relatively short data series is recorded. Using this method, the dynamical structure of the Gray-Scott model is detected from its rich spatiotemporal patterns. Finally, this method is successfully applied to the experimental data of Chua's circuit, which promises its potential of the application to realistic systems.