Data entropy decomposition into functional and uncertainty components

Abstract

Information entropy is a fundamental property of measurement data and can be decomposed into functional and relative uncertainty components. The purpose of this paper is to mathematically isolate functional entropy and to demonstrate its application to model data. The equation for functional entropy is developed and its utility is demonstrated using hypothetical data that describes different processes that contribute to the same effect. Functional entropy assigns an effectiveness value to these processes in the fundamental units of information, which makes it a useful tool for comparing different data sets and their associated processes.