Figure 10- - uploaded by Motonari Tanabu
Content may be subject to copyright.
Source publication
There are several diagram methods for data modeling like a class diagram. It is very hard to describe a big data model of a large enterprise system into one diagram. A large set of partial data models are used during designing an information system for a large enterprise. The skill of modelers makes fluctuation and discrepancy among data models. It...
Context in source publication
Citations
... Section 3.1 has appeared in[26,27,29,25,28] ...
... Section 3.2 has appeared in[26,25,28] ...
... A part of Subsection 6.3.1 has appeared in[26] ...
In recent years, the information systems of large enterprises have grown in size. Therefore , it is important to develop a data model for processing a large amount of complex information; it is also important to analyze several requirements for a large information system that is capable of handling the large amount of complex information. Designers develop many partial data models instead of a single data model of a large information system in a manner similar to functional decomposition.
However, some questions arise in the minds of experienced designers while designing partial data models in some situations. To answer these questions and to help designers to understand the need to do so, it is important to develop some theories that provide the tracing steps of sub-questions; these steps explain why the questions are answered. Some theories for data models are described algebraically because the algebra for data models consists of a domain comprising data models and operations among them. The purpose of this study is to develop the algebras of data models as theories of data modeling for answering these questions.
This thesis presents a study on the desirable algebra of class diagrams algebra for understanding the relationship between data models. Further, this thesis contributes toward proving three desirable algebras for class diagrams as Boolean initially and using these algebraic equations to answer several questions arising in the minds of designers. The algebra for the abstract notation of class diagrams helps to describe right class diagrams such as well-formed class diagrams, preventing the syntactic inconsistency in composing and decomposing of class diagrams. The algebra for predicates mapped from class diagrams, such as well-formed predicates mapped from well-formed class diagrams, as intentional semantics of the class diagrams helps to compose and decompose the right data models. The algebra for instances satisfying class diagrams as extensional semantics of class diagrams helps to efficiently implement a data model without additional codes for exchanging data between compositional data models.
... However, they haven't been clear detail of algebraic structure for a merging operation. Enjo, Tanabu and Iijima [5,6] discussed syntactical inconsistency issues. They proposed algebraic operations such as merge and difference to prevent inconsistency. ...
An enterprise service system is large and complex and manages big and complicated data. A large set of partial data models is used in designing an enterprise service system because each service application consisting of an enterprise service system uses only a part of big and complicated data in most cases. These partial data models provide several complementary views on the system to be developed. This however leads to a need for compositional models that are able to produce a single integrated model. These data models are often described by a class diagram of Unified Modeling Language because it is a very popular modeling language and describing a static view of a system. In this paper, we propose syntactical foundation of algebra on a family of well-formed class diagrams with composition operations-merge, difference, complement and intersection operations. We then show that the algebraic properties as associativity, commutativity and involutivity of this foundation are desired for model management to develop an enterprise service system.
... However, they did not defined it mathematically and it is unclear to algebraic structure. There are several other studies[2,3,7,15] discussed merge operations for model composition . Bézivin, et al.[2] discussed three model composition tools such as Atlas Model Weaver, the Epsilon Merging Language and the Glue Generator Tool which were developed in Modelware project. ...
... However, they discussed requirements for model management operations and did not mention concrete operations of model composition. Enjo, Tanabu and Iijima[7,8,9] studied model composition and discussed syntactical inconsistency issues. They proposed algebraic operations such as merge and difference in terms of first order logic for preventing inconsistency. ...
A large set of partial data models is used in designing a large information system. These partial data models provide several complementary views on the system to be developed. This however leads to a need for compositional models that are able to produce a single integrated model. These data models are often described by a class diagram of Unified Modeling Language because it is a very popular modeling language describing a static view of a system. In this paper, we present syntax and semantics of a class diagram describing a data model. We propose a family of well-formed class diagrams as a domain of class diagram algebra and composition operations as merge and difference operations. We then show that algebraic properties as associativity, commutativity and involutivity are desired for model management to develop a large information system.
