Statistical learning theory, model identification and system information content

Abstract

Statistical learning theory offers a powerful framework for uncertainty modeling and management in complex business and engineering systems. However, the implementation of this theory requires from the user (business manager or engineer) to assimilate some advanced and fairly complex mathematical concepts. Because of that, the acceptance and use of this powerful theory in business management and engineering have been very limited. This article presents an intuitive introduction to the foregoing theory using simple mathematical tools. Metaphoric images are utilized to derive the main variables that govern the uncertainty in a system, from the statistical learning theory's viewpoint. A general expression for uncertainty models is then obtained. The structure of this expression is the same as that of the uncertainty mathematical models that have been developed in statistical learning theory.