INHERITANCE COMES OF AGE - APPLYING NONMONOTONIC TECHNIQUES TO PROBLEMS IN INDUSTRY

Authors
Citation
L. Morgenstern, INHERITANCE COMES OF AGE - APPLYING NONMONOTONIC TECHNIQUES TO PROBLEMS IN INDUSTRY, Artificial intelligence, 103(1-2), 1998, pp. 237-271
Citations number
42
Categorie Soggetti
Computer Science Artificial Intelligence","Computer Science Artificial Intelligence
Journal title
ISSN journal
00043702
Volume
103
Issue
1-2
Year of publication
1998
Pages
237 - 271
Database
ISI
SICI code
0004-3702(1998)103:1-2<237:ICOA-A>2.0.ZU;2-C
Abstract
Nonmonotonic reasoning is virtually absent from industry and has been so since its inception; the result is that the field is becoming incre asingly marginalized within Al. We argue that this is largely because researchers in the area focus exclusively on commonsense problems whic h are irrelevant to industry and because few efficient algorithms or t ools have been developed. A sensible strategy is thus to focus on indu stry problems and to develop solutions within tractable subtheories of nonmonotonic logic. We examine an example of nonmonotonic reasoning i n industry-inheritance of business rules in the medical insurance doma in - and show how the paradigm of inheritance with exceptions can be e xtended to a broader and more powerful kind of nonmonotonic reasoning. This is done by introducing formula-augmented semantic networks (FANs ), semantic networks which attach well-formed formulae to nodes. The p roblem of inheriting well-formed formulae within this structure is exp lored, and an algorithm is given and discussed. Finally we discuss the underlying lessons that can be generalized to other industry problems . (C) 1998 Published by Elsevier Science B.V. All rights reserved.