A method for identifying a subset of components of a system

摘要

A method of identifying a subset of components of a system based on data obtained from the system using at least one training sample from the system is disclosed. The method comprises obtaining a linear combination of components of the system and weightings of the linear combination of components. The weightings have values based on data obtained from the at least one training sample. The at least one training sample has a known feature. A model of a probability distribution of the known feature is obtained. The model is conditional on the linear combination of components. A prior distribution for the weighting of the linear combination of the components is obtained. The prior distribution comprises a hyperprior having a high probability density close to zero. The hyperprior is such that it is not a Jeffrey's hyperprior. The prior distribution and the model are combined to generate a posterior distribution and the subset of components based on a set of the weightings that maximise the posterior distribution is identified.