Using the classification abstraction to model relational database applications - a case study

Author(s)

Alexandre Della Nina

• Document type: Master's Dissertation
• Press: São Carlos.
• Institution: Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC/SB)
• Defense date: 2000-05-19

Examining board members

Rosely Sanches; Maria Cristina Ferreira de Oliveira; Marina Teresa Pires Vieira

• Advisor: Caetano Traina Junior

Abstract

Current data modais and tools to aid in software development do not have ways to express occurrences of the Classification Abstraction. Situations in the real world where it occurs are represented through other mechanisms or even not represented at ali, introducing semantic weaknesses in the design, hardly restored in subsequent phases. Such fact are due mainly to the little support of this abstraction. One symptom of this fact is that ali widely accepted data modeling tool, like the Extended Entity_Relationship (EE-R) Model, does not have mechanisms to represents it. This work characterizes the classification abstraction and distinguishes it from the other, mainly from the generalization one. It also give to designers clues to correctly use classification in a project. The theoretical foundation of the classification and the knowledge acquired In of real case study - the development of a system for medica( cooperatives automation were used to guide this work. We propose a new extension to the EE-R Model to include a semantic constructor supporting classification. We designate this new model the Classification Entity - Relationship model (CE-R Model). Using this new model, we analyze the interactions of classification with the other existing constructors, and define the rufes to map classification constructs to the Relational Model. This is the first time in the literature of the area that this mapping is described. (AU)