Abstract: A method is disclosed for modelling a physical phenomenon that has associated spectral data. The method includes the step of obtaining the spectral data (12) on at least two samples (10) from a material, substrate or electrical activity such as electrical activity of a brain. The method includes performing an analysis of the samples of the phenomenon (15) and storing the spectral data and characteristic information in a library (14). The method includes generating from the library an individualized modelling equation as a function of the spectral data and characteristic information, wherein the modelling equation includes linear and non-linear dimensions and wherein the step of generating includes a type of kernel-learning means (17) for reducing at least the non-linear dimensions associated with the modelling equation. The method may include a validation step in which performance of the model may be assessed based on a simulation of real world use (18), ie.