Expert systems: MYCIN

It is widely accepted that probabilistic approach was first developed and applied in expert systems MYCIN and PROSPECTOR. In these expert systems the likelihood estimates are calculated for several competitive hypothesis H on the basis of available evidences E. In the expert system MYCIN each hypothesis was estimated by a confidence factor CF(H Ei,E2. . . ) as a difference of estimates for the measure of belief MB P[ Ei,E2,. .) and the measure of distrust MD P[ Ei,E2,...) ... 

The earliest practical use of an expert system was made in the software named MYCIN for diagnosing a toxic poison from the symptoms of a patient and recommending the antidote (62). This type of activity is generally carried out by a human expert who processes information about a situation (in this case, symptoms of a patient), refers to the expert s experience and expert knowledge, and then recommends action (in this case, the antidote). 

Many readers will have read about medical applications, the MYCIN and INTERNIST programs. There are many systems being developed to diagnose equipment failures. Layout and planning of manufacturing facilities are obvious applications. Chemistry and molecular biology systems were among the earliest examples of expert systems and are now embodied in commercial systems. 

An expert system shell developed in the MYCIN project is EMYCIN, which was used to develop other expert systems. One of these systems is PUEE, designed for the domain of heart disorders. Another outcome was the ventilator manager (VM) program developed as a collaborative research project between Stanford University and Pacific Medical Center in San Francisco within the scope of a Ph.D. thesis by Lawrence M. Fagan [6]. VM was designed to interpret on-line quantitative data in the intensive care unit. The system measures the patient s heart rate, blood pressure, and the status of operation of a mechanical ventilator that assists the patient s breathing. Based on this information, the system controls the ventilator and makes necessary adjustments. 

MYCIN is one of the first expert systems for medical diagnosis that supports physicists in the diagnostic process. 

Buchauau, B.G. aud Shortliffe, E.H., Eds., Rule-Based Expert Systems The MYCIN Experiments of the Stanford Heuristic Programming Project, Addisou-Wesley, Read-iug, MA, 1984. Out of priut electrouically available at http //www.aaaipress.org/Clas-sic/Buchauau/buchauau.html. 

Tools for rule-based expert systems (as well as manual methods) should evaluate the consistency and completeness of the rules. The TEIRESIAS program (Davis 1976) linked to the MYCIN infectious disease system was one of the first attempts to develop an automated verification tool. Later work by Suwa et al. (1982) for the ONCOCIN (clinical oncology) system examined a rule set as it was read into the system. This rule checker assumes that for each combination of attribute values appearing in the antecedent a corresponding rule exists. 

The problem with automated tools is that many times they are software/system dependent. For example, CHECK has been ported to the Automated Reasoning Tool (ART) framework but to little else (short of a MYCIN type structure). Thus, automated tools are not necessarily available for off-the-shelf applications in shell, language or tool expert system development. Cost-to-benefit-type trade-off studies need to be undertaken before fully automated tools are fully recognized as a (partial) means of expert system V V. 

Whereas expert system languages for reasoning like PROLOG are universally qualified (all/all not) and nonquantitative, MYCIN allows an aspect of existential qualification (some, some not) and is quantified. The rules have both a rule form as above, and a certainty factor (CF). For those of a mathematical bent, the MYCIN certainty factor is defined by the following equation ... 

In the 1960 s, Edward Feigenbaum and other scientists at Stanford University built two early expert systems DENDRAL, which classified chemicals, and MYCIN, which identified diseases. These early expert systems were cumbersome to modify because they had hard-coded rules. By 1970, the OPS expert system shell, with variable rule sets, had been released by Digital Equipment Corporation as the first commercial expert system shell. In addition to expert systems, neural networks became an important area of artificial intelligence in the 1970 s and 1980 s. Frank Rosenblatt introduced the Perceptron in 1957, but it was Perceptrons An Introduction to Computational Geometry (1969), by Minsky and Seymour Papert, and the two-volume Parallel Distributed Processing Explorations in the Microstructure of Cognition (1986),... 

The reliability of data communication elements is an important constituent part on the quality of transferred information (IQ). It may be stated that IQ consists of accessibility, actual value, completeness, credibility, flexibility, form, meaning over time, relevance, reliability, selectivity, validity (Olaisen 1990, MITIQ). All the above components of IQ tend to be encumbered by some uncertainty. Whether associated with measurement or caused by the unreliability of equipment or software. Hence, the proposal to determine IQ using uncertainty modelling. There are several major methods of uncertainty modelling. The best known are Bayes networks based on conditional probability, the theory mathematical evidence based on the Dempster-Shafer theory, the method making use of the Certainty Factor (Clf) created by Buchanan, Shortliffe for the MYCIN expert system, the theory of fuzzy sets or the theory of rough sets. 

Practically, there are many expert systems, and the most well-known and often used are Dendral, Prospector, Mycin, XCON and many others (Shim, et al. 1999 Chromiec, Strzemieczna, 1994 Hayes-Roth et al. 1983), and the basic language used to record and perform logic operations is the one used in Prolog (Bubnicki, 1993 Bratko, 2001 Puppe, 1993). 

Because of the rigorous separation between knowledge base and problem-solving strategy, MYCIN served as a starting point for one of the first tools for developing expert systems (XPS). The knowledge base of MYCIN was deleted but all... 

---