Advertisement

Abstract

Fact-Oriented Modeling is a conceptual approach to modeling and querying the information semantics of business domains in terms of the underlying facts of interest, where all facts and rules may be verbalized in language that is readily understandable by non-technical users of those business domains. Unlike Entity-Relationship (ER) modeling and UML class diagrams, fact-oriented modeling treats all facts as relationships (unary, binary, ternary etc.). How facts are grouped into structures (e.g. attribute-based entity types, classes, relation schemes, XML schemas) is considered a lower level, implementation issue that is irrelevant to the capturing essential business semantics. Avoiding attributes in the base model enhances semantic stability and populatability, as well as facilitating natural verbalization. For information modeling, fact-oriented graphical notations are typically far more expressive than those provided by other notations. Fact-oriented textual languages are based on formal subsets of native languages, so are easier to understand by business people than technical languages like OCL. Fact-oriented modeling includes procedures for mapping to attribute-based structures, so may also be used to front-end other approaches.

Though less well known than ER and object-oriented approaches, fact- oriented modeling has been used successfully in industry for over 30 years, and is taught in universities around the world. The fact-oriented modeling approach comprises a family of closely related “dialects”, the most well known being Object-Role Modeling (ORM), Natural Language Information Analysis Method (NIAM), and Fully-Communication Oriented Information Modeling (FCO-IM). Though adopting a different graphical notation, the Object-oriented Systems Model (OSM) is a close relative, with its attribute-free philosophy.

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Terry Halpin
    • 1
  • Robert Meersman
    • 2
  1. 1.Neumont University 
  2. 2.Vrije Universiteit Brussel 

Personalised recommendations