Article ID Journal Published Year Pages File Type
378769 Data & Knowledge Engineering 2015 24 Pages PDF
Abstract

AsbtractSoftware interoperability and application integration can be realized through using their respective conceptual data models, which may be represented in different conceptual data modeling languages. Such modeling languages seem similar, yet are known to be distinct. Several translations between subsets of the languages' features exist, but there is no unifying framework that respects most language features of the static structural components and constraints. We aim to fill this gap. To this end, we designed a common and unified ontology-driven metamodel of the static, structural components, and constraints in such a way that it unifies ER, EER, UML Class Diagrams v2.4.1, and ORM and ORM2 such that each one is a proper fragment of the consistent metamodel. The paper also presents some notable insights into the relatively few common entities and constraints, an analysis on roles, relationships, and attributes, and other modeling motivations are discussed. We describe two practical use cases of the metamodel, being a quantitative assessment of the entities of 30 models in ER/EER, UML, and ORM/ORM2, and a qualitative evaluation of inter-model assertions.

Related Topics
Physical Sciences and Engineering Computer Science Artificial Intelligence
Authors
, ,