Automating the knowledge acquisition process in medical expert systems (original) (raw)

2005

M. AL ABDULSALAM L'intelligence économique coopérative. Une approche par les réseaux . . . . . . . . . . . . . . . . . . . . . . . . . 11 ... B. GAY, B. DOUSSET Cartographie de réseaux d'alliances et analyse stratégique . . . . . . . . . . 37 ... F. AMARDEILH, T. FRANCART Enrichissement de bases de connaissances par l'annotation sémantique. Plate-forme web sémantique couplée avec des outils linguistiques pour des activités de veille et d'édition . . . . . . . . . . . . . 53 ... N. USUNIER, M.-R. AMINI, P. GALLINARI Résumé automatique de texte avec un algorithme d' ...

Towards the principled engineering of knowledge

AI Magazine, 1982

The acquisition of expert knowledge is fundamental to t,he cre;ltion of expert syst,ems The conventional ~pp1oach to building expert syst,ems assumes that the knowlcdgc exists, and that iL is feasible to find 2x1 cxpert who has the knowledge and can articulat,e it in rollaboraLion with a knowlcdgc engineer This al title considers i.he practice of knowledge enginerling when t,hesc asslm~ptions can not bc strictly jusi,ified It tllaws on our cxpcricnccs in the design of VLSI design methods, and in the prototyping of an expert assistant fol VI31 design WC suggest. rncthods for expanding the practice of knowledge engineering when applied to fields t.hat ale fragmrnt,ed and undergoing rapid evolution We ouf,line how the expanded practice can shape and accelerate the process of knowlcdgc generation and refinement Ollr examples also clarify some of (.he una~ ticulatcd present pl ad.ice of knowledge cnginccring Thanks cspccially to Daniel Boblow for helping us to discover, refine, and al ticulate many of these ideas We are also grateful to John Se+ Brown, Douglas Lcnat, C:hrist.opher Tong, and Michael Williams for their t.horough reviews of drafts of this article Thanks also to the n~emhe~s of t,he KBVLSI project: Alan Bell, IIarry Uarrow, Daniel Hohrow, ilarold Brown, Phil Gerring, Gordon Foystel, Cordon Novak, Christopher Tong, and Narindcr Singh, who have participated in the knowlcdgc cnginccl ing and expel t systems aspec1.s of t.he I."oject The synergist.ic cornhinxf,ion of their contributions to the project, merging idcns from a wide variety of viewpoints, has given us all a sense of excitcmcnt and common pul pose Thanks to the Xerox CTorporation for providing the intellectual and computat.ional environment, in which this work could Ix done This research is hcing rondllcted as pal t of the KBVLSI project, in collabora-Con with the Hcuristir Programming Project at Stanfold University The Stanfold component of the research is funded by the Defense Adva uced Research PI oje(%s Agency A STRONG MOTIVATION