Introduction to the notion
Semantic modeling consists in identifying the objects and concepts that come within the domain of study, identifying their common characteristics and organizing them in the most economical way possible.
The class is the basic unit of semantic modeling. To avoid all ambiguity, we will call it the “semantic class”. The semantic class enables all the semantics connected to a real object, a set of similar objects or a concept to be restored. Candidate classes are classes that the modeler is considering including in the model, to capture part of the meaning. They often correspond to a substantive used in speech.
The scope of semantic modeling is circumscribed according to the pertinence criterion, that is to say, by relating this effort to an explicit objective. Yet, it is often necessary to go slightly beyond this objective, especially if it is a project objective. Indeed, to get to the “right” representation more quickly, the modeler has to anticipate future developments or uses of the model.
Another question that the modeler encounters is that of the level of genericity or abstraction of the model. On the one hand, the genericity of the model increases it scope and therefore constitutes an advantage. On the other hand, it complicates the modeler’s task and makes it more difficult to communicate the model. This question can only be decided one way or another by analyzing the context and ambition. It implies an architecture vision.