To read the full version of this content please select one of the options below:

On the construction of selection systems

Michael Buckland (Professor in the School of Library and Information Studies, University of California, Berkeley)
Christian Plaunt (Doctoral student in the School of Library and Information Studies, University of California, Berkeley)

Library Hi Tech

ISSN: 0737-8831

Article publication date: 1 April 1994

123

Abstract

This article examines the structure and components of information storage and retrieval systems and information filtering systems. Analysis of the tasks performed in such selection systems leads to the identification of 13 components. Eight are necessarily present in all such systems, mechanized or not; the others may, but need not be, present. The authors argue that all selection systems can be represented in terms of combinations of these components. The components are of only two types: representations of data objects and functions that operate on them. Further, the functional components, or rules, reduce to two basic types: 1) transformation, making or modifying the members of a set of representations, and 2) sorting or partitioning. The representational transformations may be in the form of copies, excerpts, descriptions, abstractions, or mere identifying references. By partitioning, we mean dividing a set of objects by using matching, sorting, ranking, selecting, and other logically equivalent operations. The typical multiplicity of knowledge sources and of system vocabularies is noted. Some of the implications for the study, use, and design of information storage and retrieval systems are discussed.

Citation

Buckland, M. and Plaunt, C. (1994), "On the construction of selection systems", Library Hi Tech, Vol. 12 No. 4, pp. 15-28. https://doi.org/10.1108/eb047934

Publisher

:

MCB UP Ltd

Copyright © 1994, MCB UP Limited

Related articles