Browse Prior Art Database

MULTIPLE VIEWS AND ABSTRACTIONS WITH AN EXTENDED ENTITY-RELATIONSHIP MODEL

IP.com Disclosure Number: IPCOM000128154D
Original Publication Date: 1979-Dec-31
Included in the Prior Art Database: 2005-Sep-15
Document File: 12 page(s) / 44K

Publishing Venue

Software Patent Institute

Related People

Gerd Schiffner: AUTHOR [+4]

Abstract

The past few years have seen a growing awareness of the fact that a conceptual framework, involving a three schema technology, is essential for the develcpw.ent of data base systems. The novel element in this frame-work, compared to previous database technology, is the emergence of the conceRtual schema: a set of rules describing the.information relevant to a given enterprise. Such a description should encompass object types and roles/ relations, as well as consistency and dependency constraints (1,4,12,131. How this information is presented to a,program by various users is described in the external schema, while the internal schema describes how to represent it in storage (still one step removed from actual physical records, but including physical sequencing, selection of indices, etc.). There is also the need for two mappings, one between the conceptual schema and the internal schema, the other between the conceptual schema and the external schema. Since the process of defining the relevant information to an enterprise is closely related to the system analysis and design activity, the conceptual schema is an important tool in the communication between end-users and managers on one hand and database designers and implementers on the other hand. Additional advantages of the conceptual schema concept are improved semantic control over the data and a higher-dearee of data-independence, due to the relative stability of the conceptual schema (2,7,191. Although there is not yet a consensus about what constitutes the best set of concepts to be incorporated in a conceptual schema, a number of models have been identified as possible candidates [3]. Among these, the Entity- Pelationship (E-R) model proposed by Chen [61 has many of these desired features; it is easy to formulate, it is easy to understand, and in addition it includes a concise diagramatic technique which can be used in the process of system analysis [6,7,101. In addition, the simplicity of the E-R model and its semantic structure, closely resembling that of sentences in natural languages, also make it a desirable choice for the external schema. However, to use the E-R model in this dual role, it is necessary to expand its capa-bility to support multiple views and abstractions, along the lines suggested by Smith and Smith [15,161. 'In this paper we first describe some general properties of the E-R model. Next, we discuss how to extend the model in order to support more abstractions, (views) and correspondingly augment the diagrammatic tech-niq!ae. To realize these concepts at the external (user) level we present a suitable data definition language. Finally, in the last section we give some guidelines for translating the Extended-Entity-Relationship (E-E-R) diagrams at the internal level in to an executable DBTG network schema.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 9% of the total text.

Page 1 of 12

THIS DOCUMENT IS AN APPROXIMATE REPRESENTATION OF THE ORIGINAL.

MULTIPLE VIEWS AND ABSTRACTIONS WITH AN EXTENDED ENTITY- RELATIONSHIP MODEL

Gerd Schiffner and Peter Scheuermann Department of Electrical Engineering and Computer Science Northwestern University Evanston, IL 60201

Abstract

The features of the Entity-Relationship model are examined, as a means for representing both the conceptual schema and the external schema in a database system. In order to serve in this dual capacity, its modelling features are extended to support multiple user views. We show how to represent model and present a suitable diagrammatic technique. To represent these concepts at the user level we propose a data defini-tion language with a concise and simple structure. Furt~e rmore, guide-lines are given for interfacing the E-E-R schema with existing DBTG-like systems.

Key Words: conceptual schema, external schema, database systlems, data definition language, data abstraction.

This work is sponsored in part by NSF Grant 1,1.,ICS77-03904; also, to appear in the journal of Computer Languages.

1. Introduction

The past few years have seen a growing awareness of the fact that a conceptual framework, involving a three schema technology, is essential for the develcpw.ent of data base systems. The novel element in this frame-work, compared to previous database technology, is the emergence of the conceRtual schema: a set of rules describing the.information relevant to a given enterprise. Such a description should encompass object types and roles/ relations, as well as consistency and dependency constraints (1,4,12,131. How this information is presented to a,program by various users is described in the external schema, while the internal schema describes how to represent it in storage (still one step removed from actual physical records, but including physical sequencing, selection of indices, etc.). There is also the need for two mappings, one between the conceptual schema and the internal schema, the other between the conceptual schema and the external schema. Since the process of defining the relevant information to an enterprise is closely related to the system analysis and design activity, the conceptual schema is an important tool in the communication between end-users and managers on one hand and database designers and implementers on the other hand. Additional advantages of the conceptual schema concept are improved semantic control over the data and a higher-dearee of data-independence, due to the relative stability of the conceptual schema (2,7,191.

Although there is not yet a consensus about what constitutes the best set of concepts to be incorporated in a conceptual schema, a number of models have been identified as possible candidates [3]. Among these, the Entity- Pelationship (E-R) model proposed by Chen [61 has many of these desired features; it is easy to formulate, it is easy to understand, and in addition it includes a concise diagramatic techn...