CumInCAD is a Cumulative Index about publications in Computer Aided Architectural Design
supported by the sibling associations ACADIA, CAADRIA, eCAADe, SIGraDi, ASCAAD and CAAD futures

PDF papers
References
id ddss9463
authors Lucardie, Larry
year 1994
title A Functional Framework For Conceptual Modelling
source Second Design and Decision Support Systems in Architecture & Urban Planning (Vaals, the Netherlands), August 15-19, 1994
summary A conceptual model is not only indispensable for the design and implementation of knowledge based systems, but also for their validation, modification, maintenance and enhancement. Experience indicates, however, that in many cases reality is not well reflected in a full-fledged conceptual model. What is systematically lacking in the prevailing conceptualization methods is a well-developed theory of knowledge that underlies conceptualization methods: a theory that precedes the process of forming meaningful classifications and that precedes the specification of a conceptual model. To date, conceptualization methods are based on the probabilistic assumption that, in essence, all conditions necessary for creating a classification, are provided initially and can easily be revealed by utilizing mathematical measures of similarity. Another frequently occurring prototypical assumption is that for creating a classification, necessary conditions are sufficient. Furthermore, it is assumed that the categories of conditions are a priori fixed and unconditional. That conceptualizing takes place without any explicit background knowledge about goals of classifications and without contextual influences and that categorizations have an unconditional status are not viewed as problems. In contrast to these approaches, the functional view states that relevant descriptive attributes are not necessarily a priori given but should be acquired through knowledge about goals of classifications and about contexts. It is also asserted that an explicit concern for necessary conditions will not suffice for capturing the dynamics of reality. Furthermore, the functional view puts forward that a goal- and context-oriented strategy leads to the reconstruction of new attributes and categorizations with a dynamic status. The aim of this paper is to discuss the theoretic and practical merits of the functional view compared to the probabilistic and prototype approaches. Conceptual models developed in the Computer Integrated Manufacturing-Project will serve as illustrations for the main ideas.
series DDSS
email
references Content-type: text/plain
last changed 2003/08/07 16:36
pick and add to favorite papersHOMELOGIN (you are user _anon_636514 from group guest) CUMINCAD Papers Powered by SciX Open Publishing Services 1.002