A method for optimization of a conceptual model

Abstract

Optimization of a conceptual model is a non-trivial task. A set of rules for determining the "best" one out of a number of candidate solutions is introduced. A key point is the definition of a set of quantifiable characteristics of a conceptual model; a measurement of the characteristics together with a fixed measurement of the decision-maker's preferences are used to quantify the quality of the candidate solutions relative to each other. The chosen characteristics are: size, change pr. month, data description inaccuracy, semantic relevance, semantic inaccuracy and l/0-model size.