Browse Prior Art Database

Subject Selection Using a Table of Possible Relationships

IP.com Disclosure Number: IPCOM000062334D
Original Publication Date: 1986-Nov-01
Included in the Prior Art Database: 2005-Mar-09
Document File: 4 page(s) / 45K

Publishing Venue

IBM

Related People

Woods, D: AUTHOR

Abstract

This invention relates to a method for operating a digital computer referencing a data base to derive from an explicitly selected subject a subset of subjects from data bases (indexed in, for example, the IBM Data Dictionary) needed to be selected for further processing based on the relationships of the subset to other previously selected subjects in the data base. This method may be applied to any collection of data that can be characterized in terms of types of data entities and relationships between those types in a data base environment, such as that provided by the IBM Information Management System (IMS). Each subject in the IBM Data Dictionary, hereinafter referred to as "the Dictionary", is an instance of a particular category of subjects.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 36% of the total text.

Page 1 of 4

Subject Selection Using a Table of Possible Relationships

This invention relates to a method for operating a digital computer referencing a data base to derive from an explicitly selected subject a subset of subjects from data bases (indexed in, for example, the IBM Data Dictionary) needed to be selected for further processing based on the relationships of the subset to other previously selected subjects in the data base. This method may be applied to any collection of data that can be characterized in terms of types of data entities and relationships between those types in a data base environment, such as that provided by the IBM Information Management System (IMS). Each subject in the IBM Data Dictionary, hereinafter referred to as "the Dictionary", is an instance of a particular category of subjects. Relationships can be established between any two categories (both categories could be the same) with each relationship being uniquely defined by the two categories (leftcat relkw rightcat) will be referred to as a relationship type or reltype in the following text. The overall "data structure" of an installation's Data Dictionary (i.e., the combination of the subject categories and reltypes that are defined in the Dictionary) might be represented pictorially in terms of a graph or diagram in which each subject category is represented by a "node", and the reltypes defined between the different categories are represented by lines connecting the different nodes. Relationship types that have a sense of direction might be represented by lines that are arrows pointing from the superior category to the subordinate category, while reltypes with no such sense of direction would be represented simply by lines without such a directional marking. (Relationship types that are defined between subjects in the same category would be represented by lines or arrows that loop out of and back to the same node.) The relationship PARAGRAPH/CONTAINS/SENTENCE (leftcat = PARAGRAPH, relkw = CONTAINS, rightcat = SENTENCE) and the relationship SENTENCE/CONTAINED-BY/PARAGRAPH (leftcat = SENTENCE, relkw = CONTAINED- IN, rightcat = PARAGRAPH) are examples of reltypes that are inverse versions of each other that could be used to process the same line or arrow from opposite directions. With this representation, a Dictionary "Structure" might be visualized as simply a piece of the overall data structure--such as one would obtain by "coloring" some subset of the category nodes, together with some of the connecting reltype lines. Several points should be noted with respect to the contents of a Dictionary Structure: 1. A Structure need not be a pure hierarchy. A Structure may have multiple root nodes, and may contain loops or knots (i.e., multiple reltype paths connecting the same pairs of categories). 2. A Structure need not contain all of the reltypes that have been defined between the different categories represented in the Structure. Indeed, one of the purposes of...