TEXT ANALYSIS FOR CONSTRUCTING DESIGN REPRESENTATIONS (REPRINTED FROMARTIFICIAL-INTELLIGENCE IN DESIGN, PG 21-38, 1996)

An emerging model in concurrent product design and manufacturing is th e federation of workgroups across traditional functional 'silos'. Alon g with the benefits of this concurrency comes the complexity of sharin g and accessing design information. The primary challenge in sharing d esign information across functional workgroups lies in reducing the co mplex expressions of associations between design elements. Collaborati ve design systems have addressed this problem from the perspective of formalizing a shared ontology or product model. We share the perspecti ve that the design model and ontology are an expression of the 'meanin g' of the design and provide a means by which information sharing in d esign may be achieved. However, in many design cases, formalizing an o ntology before the design begins, establishing the knowledge sharing a greements or mapping out the design hierarchy is potentially more expe nsive than the design itself. This paper introduces a technique for in ducing a representation of the design based upon the syntactic pattern s contained in the corpus of design documents. The association between the design and the representation for the design is captured by basin g the representation on terminological patterns at the design text. In the first stage, we create a 'dictionary' of noun-phrases found in th e text corpus based upon a measurement of the content carrying power o f the phrase. In the second stage, we cluster the words to discover in ter-term dependencies and build a Bayesian belief network which descri bes a conceptual hierarchy specific to the domain of the design. We in tegrate the design document learning system with an agent-based collab orative design system for fetching design information based on our 'sm art drawings paradigm. (C) 1996 Kluwer Academic Publishers, Published by Elsevier Science Ltd.