E. H. Shortliffe, "Computer-Based Medical Consultations: MYCIN," Elsevier/North Holland, New York NY, 1976.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....transferable belief model [87, 90] Another much studied approach is possibility theory [26, 103] which grew out of work on fuzzy sets [102] There are numerous other numerical techniques for dealing with uncertainty often developed from pragmatic considerations. These include certainty factors [86], probabilistic logic [63] and belief intervals [21] to name but a few. The methods that we shall consider in the following sections are the main formal theories introduced to handle uncertainty probability theory, possibility theory, and evidence theory. For theories that have traditionally ....

....consider the application of fuzzy techniques to databases. 2.5 Other approaches There are a number of other numerical techniques which, although we do not have space to consider them in any detail, are worth mentioning for their particular historical or theoretical interest. Certainty factors [86], perhaps because of their simplicity and intuitive appeal, have been widely used to handle uncertainty. The certainty factor approach assigns a numerical weight, the certainty factor, to the consequent of every If hevidencei then hhypothesisi rule in a rule based system. The value of the ....

E. H. Shortliffe. Computer-Based Medical Consultations: MYCIN. Elsevier, New York, NY, 1976.

....that preserve modularity of processing and linguistic knowledge, hence making the system transparent and reusable for various applications. Production systems have been used both for modeling human thought (e.g. Newell, 1973] and for the con struction of knowledge based expert systems (e.g. [Shortliffe, 1976]) In spite of the modu larity gained by separating the rule basis from the interpreter, production systems have disappeared from the focus of current research because of their limited transparency caused by various types of side effects. In particular, side effects could modify the data base in ....

E. H. Shortliffe. Computer- based Medical Consultations: MYCIN. El- sevier, New York, 1976. 11o

....applicability condition examines one element of the current instructional context, from the student model, the domain representation, or the instructional plan. The individual conditions are combined to determine a final score for the PMM, using a calculus akin to MYCIN s certainty factors ([Shortliffe 1976]) For example, one applicability condition states that less student knowledge about a domain concept can provide evidence for the PMM s greater applicability, and more knowledge implies less applicability. A PMM s removal conditions provide a facility for determining when the misrepresentation ....

Shortliffe, E.H. Computer-Based Medical Consultation: MYCIN. New York: Elsevier.

....operations which are continuous (in some reasonable sense) these two approaches coincide. 1 Introduction: The Problem of Combining Evidence In the development of expert systems the problem of combining multiple pieces of evidence about a hypothesis H was recognized early as an important issue [1, 12]. Simply stated, the problem is as follows: If two independent pieces of evidence support the hypothesis H , A with degree a, and B with degree b, then what is our resulting degree of belief in H If these degrees a and b are the only information we have, then the resulting resulting degree of ....

....i.e. we should have a b a; on the other hand, if we choose a t conorm, we get a b a. Thus, for combining pieces of evidence, we need operations which are more general than t conorms. First such operations were used already in the historically first expert system MYCIN (see, e.g. [1, 12]) In this system, degrees of belief take values from the interval [ Gamma1; 1] instead of the more traditional [0; 1] Here, negative values describe negative evidence, and positive values describe positive evidence. As we move from Gamma1 to 1, we go from the evidence which absolutely 100 ....

E. H. Shortliffe, Computer-based medical consultation: MYCIN, Elsevier, New York, 1976.

....the Stanford Research Institute in 1972, identified minerals in samples for predicting the presence of molybdenum deposits. Duda et al. 1978] Medical diagnostics MYCIN, created by Stanford in 1972, used the results of blood tests to diagnose blood diseases and recommend appropriate treatments [Shortliffe 1976]. The University of Pittsburgh produced Caduceus in 1975 to perform internal medicine diagnostics [Rolston 1988] Puff was capable of pulmonary disease diagnostics in the 1980s; the Casnet and Internist expert systems to assist in medical diagnostics were also developed in the 1980s [Matin and ....

E. H. Shortliffe. Computer-Based Medical Consultation: MYCIN. Elsevier North, Holland. 1976.

....not embody an explicit model of structure and behaviour, but rather consisted of encoded human expertise in solving particular problems in the underlying domain. Currently, the term empirical associations is often employed to denote such knowledge. The classical example of such a system is MYCIN [Shortliffe, 1976]. The model based approach to diagnosis has been successfully applied to fault finding in electronic circuits. Early work in this field is described in [Brown et al. 1982] Davis, 1984] Genesereth, 1984] and [De Kleer, 1976] The study of simple electronic circuits has yielded much insight ....

....goal of AC diagnosis, however, remains the classification of observed findings in terms of abnormality. AC diagnosis is often referred to in the literature as heuristic classification [Clancey, 1985] although this term is broader, since it also includes a reasoning strategy. The MYCIN system, [Shortliffe, 1976], is the classical system in which this conceptual approach to diagnosis has been adopted. AC diagnosis can be characterised in terms of logical deduction in a straightforward way. We shall refer to this formalisation of AC diagnosis as hypothetico deductive diagnosis. A comparison of the three ....

[Article contains additional citation context not shown here]

E.H. Shortliffe (1976). Computer-based Medical Consultations: MYCIN. New York: Elsevier.

....the probabilities of several explanations. But starting in the 1970 s, scientists involved in computer based reasoning began to recognize the potential benefits of applying symbolic reasoning techniques in clinical domains [33] The first medical reasoning program, known as the MYCIN System [32], adopted symbolic processing techniques largely in response to a conviction that computer based consultation systems, in order to be accepted by physicians, should be able to explain how and why a particular conclusion had been derived [32] Subsequently, a series of additional medical ....

.... medical reasoning program, known as the MYCIN System [32] adopted symbolic processing techniques largely in response to a conviction that computer based consultation systems, in order to be accepted by physicians, should be able to explain how and why a particular conclusion had been derived [32]. Subsequently, a series of additional medical application programs have been created. For example, the PIP [28] Present Illness Program) is designed to simulate human clinical cognition and the INTERNIST [27] diagnostic system of Pople and Myers is a computerized diagnostic program which ....

[Article contains additional citation context not shown here]

E. H. Shortliffe. Computer-based Medical Consultations: MYCIN. Elsevier, 1976.

....the search. Thus the agent may elect to search the FOX site once, because its expected response time is 60 seconds, or it may elect to search the Night Owl site many times. The choice in this case is driven by a distribution expressing expected quality or simply a quality factor similar to Mycin s[3] certainty factors. If the agent is given 95 seconds instead of 60 seconds, the choices are to search the Fan Club site once, search the FOX site once and the Night Owl Site a couple of times, or to search the Night Owl Site many times. In addition to basic WWW information servers, such as the ....

E.H. Shortliffe, Computer Based Medical Consultations: MYCIN, New York, American Elsevier.

.... The most prominent of them are: INTERNIST [3] a diagnostic aid that combines a large database of disease manifestation associations with techniques for problem formulation; EXPERT [4] and CASNET Glaucoma [5] which use physiological models for the diagnosis and treatment of eye disease; MYCIN [6], a rule based program for diagnosis and therapy selection for infectious diseases; the Digitalis Therapy Advisor [7] which aids the physician in prescribing the right dose of the drug digitalis and also explains its actions; and ABEL [8] a program that uses multi level pathophysiologic models ....

Shortliffe EH. Computer Based Medical Consultations: MYCIN. Elsevier, NorthHolland Inc., 1976.

....different scales, we can say that 0.9 and 0.8 represent the same degree of certainty in two different scales. Some scales are different even in the fact that they use an interval different from [0,1] to represent uncertainty. For example, the famous MYCIN system uses [ Gamma1; 1] Shortliffe 1976 [58], Buchanan 1984 [9] In some sense all scales are equal, but some are more reasonable than others From a mathematical viewpoint, one can use any scale, but from the practical viewpoint some of them will be more reasonable to use, and some of them less reasonable. We ll consider only practically ....

E. H. Shortliffe. Computer-based medical consultation: MYCIN (Elsevier, New York, 1976).

.... more general conditions, every or can be approximated within any given accuracy by an operation which is isomorphic to a b; see, e.g. 20] It is worth mentioning that while in fuzzy logic, we normally only use non negative degrees of belief, in the first expert systems such as MYCIN [2, 24], negative values were used as well, so our use of negative values follows a natural idea. 5 Third Step: The And Operation When describing the or operation, we implicitly assumed that all the statements come from reliable experts. In reality, different experts are reliable to different ....

E. H. Shortliffe, Computer-based medical consultation: MYCIN, Elsevier, New York, 1976.

....first in USA and later also in Europe, started in the 1970 s to work with expert systems, knowledge representations, reasoning methods, uncertainty management and models of decision making, among others. The early work resulted in many well known prototypical medical expert systems, such as MYCIN [Shortliffe 1976] and INTERNIST [Miller et al. 1982] These rule and frame based systems demonstrated that domain specific knowledge could to some extent be captured and represented, but still there remained problems such as that of tacit knowledge. As these systems were based on shallow rules, managing only ....

Shortliffe EH, Computer-based medical consultations: MYCIN. Elsevier Publishing Company, USA, 1976.

....use of wide spectrum antibiotics to cover several leading hypotheses means that misdiagnosis of the infecting organism does not necessarily lead to inadequate treatment. Careful examination of the results of a comparison of CFs with probabilistic inference presented in the original paper on CFs [135] shows that, on average, the CF based system underresponded to the diagnosticity of the data by a factor of two [159, 62] In 25 of the cases, it actually responded to the wrong direction, interpreting evidence that overall supported a conclusion to be disconfirming, or vice versa. Investigators ....

E.H. Shortliffe. Computer-based Medical Consultations: MYCIN. North Holland, New York, 1976.

....such as heuristic classification, with tasks, such as diagnosis (Clancey [1] McDermott [6] However, successful engineering requires a more fine grained analysis of domain and task characteristics. This is nowhere more true than for diagnostic systems. Since the MYCIN experiments (Shortliffe [8]) many different heuristic, model based, and hybrid approaches have been reported. For review articles see (Hayes Roth [3] and relevant chapters of (Shrobe [9] In order to understand how domain and task characteristics motivate particular system functions and architectures, this paper ....

....representation of uncertainty; modifies the search algorithm; and gives users more visibility into and control over the diagnostic process. Within TEST , failure modes may be assigned confidence factors (CF) on the basis of evidential or uncertain rules. Following the MYCIN approach (Shortliffe [8]) these rules contribute a measure of belief (MB) or disbelief (MD) to a hypothesis (failure mode) with which they are associated. TEST allows rules to have Boolean antecedents which may refer to known data or evoke questions and tests. Unlike MYCIN, TEST links rules into the semantic net of ....

Shortliffe, E., Computer-Based Medical Consultation: Mycin, Elsevier (1976).

....additional rules must be added such as Some birds fly or Ostriches don t fly. These additional rules can add significant complexity to a knowledge base. 3 Bayesian Networks (BNs) Approaches to dealing with uncertainty include fuzzy logic [40] cost based techniques [7] certainty factors [34, 35], Dempster Shafer theory [32] and probabilistic methods [27] These approaches can be used both extensionally and intensionally. Extensional systems, such as rule based systems, attach some sort of truth value to each rule or formula. The truth value for some formula is calculated functionally ....

E. H. Shortliffe. Computer-Based Medical Consultation: MYCIN. Elsevier, 1976.

.... approaches to reasoning with uncertain and incomplete information require an extensive statistical knowledge (e.g. prior and posterior probabilities for Bayesian methods, basic probability assignment in Dempster Shafer theory [3, 4, 29] prior beliefs for computing certainty factors in MYCIN [10, 31], partial probability assignments for Nilsson s Probabilistic Logic [23] In many practical situations such statistical data are either only partially available or insufficiently reliable. These circumstances raise an important question of how reasoning with incomplete information can be ....

Shortliffe, E. H. Computer-Based Medical Consultation: MYCIN (Elsevier, 1976).

.... approaches to reasoning with uncertain and incomplete information require an extensive statistical knowledge (e.g. prior and posterior probabilities for Bayesian methods, basic probability assignment in Dempster Shafer theory [15, 16, 77] prior beliefs for computing certainty factors in MYCIN [31, 80], partial probability assignments for Nilsson s Probabilistic Logic [60] In many practical situations such statistical data are either only partially available or unsatisfactorily reliable. These circumstances raise an important question of how reasoning with incomplete information can be ....

Shortliffe, E. H., 1976, Computer-Based Medical Consultation: MYCIN (Elsevier).

No context found.

E. H. Shortliffe, "Computer-Based Medical Consultations: MYCIN," Elsevier/North Holland, New York NY, 1976.

No context found.

E. H. Shortliffe, "Computer-Based Medical Consultations: MYCIN," Elsevier/North Holland, New York NY, 1976.

No context found.

E. Shortliffe. Computer-based Medical Consultations: MYCIN. Elsevier, 1976.

No context found.

E. H. Shortliffe. Computer-Based Medical Consultations: MYCIN. American Elsevier, NY, 1976.

No context found.

E. H. Shortliffe, Computer-based medical consultation: MYCIN, Elsevier, New York, 1976.

No context found.

E.H. Shortliffe. Computer-Based Medical Consultations: MYCIN. Elsevier/North-Holland, Amsterdam, London, New York (1976).

No context found.

Shortliffe EH. Computer-Based Medical Consultations: MYCIN. New York: American Elsevier; 1976.

No context found.

Shortliffe, E. H., Computer-Based Medical Consultation: MYCIN, Elsevier, New York, 1976.

First 50 documents  Next 50

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