J. Fox. On the soundness and safety of expert systems. Artificial Intelligence Medicine, 5:159--179, 1993.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....mistakes are tolerated. This does not appear to be so for expert systems. An expert system that offers advice for clinical intervention, albeit as advice intended for decision support such that the final responsibility for decision remains with the clinician, is expected to be infallible [39]. The concept of an interpretation support expert system as a distinct class of expert system, that is less interventionist than a traditional decision support expert system, has been introduced in this work. The design of the expert system as an interpretation support system is believed to have ....

J. Fox. On the soundness and safety of expert systems. Artificial Intelligence Medicine, 5:159--179, 1993.

.... in the following areas: temporal abstraction [Aben, 1995, chapter 6] problem solving methods in diagnosis [Aben, 1995, chapter 8] a simple scheduling method [Balder et al. 1993] analysing underwater sonar data [Wols, 1993, van tHolt, 1993] and safety specification in medical applications [Fox, 1993]. In [Ruiz et al. 1994] and [Aben, 1995, chapter 8] we have analysed the advantages that the use of a formal specification framework as presented here has brought to the modelling of a scheduling algorithm and of a set of diagnostic methods. In [Aben, 1995] we have taken a subset of the problem ....

J. Fox. On the soundness and safety of expert systems. Artificial Intelligence in Medicine, 5:159--179, 1993.

....stable, the time is ripe for proving the value of these languages in real life applications. Areas where KBS are already used and which seem appropriate for the use of these formal techniques are the process industry (e.g. chemical plants, energy production) health (e.g. medical decision making [29]) transport and finance. Acknowledgements We are grateful to Jan Treur, Ameen Abu Hanna, Pascal van Eck and two anonymous referees for their comments on this paper. ....

J. Fox. On the soundness and safety of expert systems. Artificial Intelligence in Medicine, 5:159--179, 1993.

....of software and knowledge bases may not be sufficient to ensure that the advice given or actions recommended in complex situations will always be appropriate. This section addresses this difficulty by proposing that systems should be able to explicitly anticipate possible hazards at runtime [Fox93, HHDW94]. The safety of a decision support system [Das95a] is the property that any actions recommended by the system will have minimal undesirable consequences. Such consequences may result from recommendations which have arisen in any of the following situations: Group I (a) Hardware failure. b) Human ....

....decision theory, while well understood, addresses a restricted range of the functions required for intelligent systems, while knowledge based systems are versatile but rather ad hoc and poorly understood. However, soundness is a necessary but not sufficient condition for safe decision making [Fox93]. Even if an R 2 L decision engine is properly specified, formally verified and correctly implemented it may still give advice which is wrong and even unsafe. Consequently the basic domino structure and R 2 L formalism incorporate constructs for explicitly representing safety knowledge, viz ....

J. Fox. On the soundness and safety of expert systems. Artificial Intelligence in Medicine, 5:159--179, 1993.

No context found.

J. Fox. On the soundness and safety of expert systems. Artificial Intelligence Medicine, 5:159--179, 1993.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC