Definition, expert system

Following this, the authors have investigated the problem of decision support system for transportation means maintenance process performance by development of DSS conception (see e.g. (Bojda Werbinska-Wojciechowska 2012)), expert system definition (see e.g. (Nowakowski Werbinska-Wojciechowska 2012a)), data uncertainty analysis (see e.g. (Nowakowski Werbinska-Wojciechowska 2012b)), or initial analysis of... 

Definition / An expert system is a computer program that manipulates large amounts of symboHc knowledge using quaUtative techniques, to solve problems that can otherwise be solved only by expert human problem solvers. Expert systems capture the human problem solver s expertise in the form of domain-specific knowledge and domain-independent problem-solving strategies. 

Definition 2 is phrased in terms of knowledge-based systems rather than expert systems. No reference is made to expert human problem solvers. Definition 2 captures the sense that the representation and manipulation of knowledge is the source of such a system s power, whether or not that knowledge is dkecdy eHcited from a human expert. 

The syntax for implementing this in an actual expert system is, of course, quite diverse and it is beyond the scope of this book to describe this in more detail. As an example the above-mentioned mle is translated in a realistic syntax. This starts by the definition of the attributes and their possible values. In this examples two attributes must be defined ... 

The next few steps are very similar to those required in any software project. One of the first stages is the clear definition of the knowledge domain. It must be clear which problems the expert system must solve. It is at this stage not the intention to define how this can be done. Clarity and specificity must be the major guides here. Fuzziness at this stage will, more than in classical software projects, have to be paid for later when different interpretations cause misunderstandings. Equally important is the clear definition of the end user(s). An expert system set up as decision support tool for professionals is totally different from an expert system that can be used as a training support for less professional people. 

Now that the definition of a volatile liquid has been settled, the expert system could apply the rule. However, this approach is clearly unsatisfactory. The all-or-nothing crisp set that defines "volatile" does not allow for degrees of volatility. This conflicts with our common sense notion of volatility as a description, which changes smoothly from low-boiling liquids, like diethyl ether (boiling point = 34.6°C), which are widely accepted to be volatile, to materials like graphite or steel that are nonvolatile. If a human expert used the rule ... 

Some Areas of Application. I next summarize some areas of application where expert systems exist or are being developed, usually by several laboratories. Some of these areas are covered in detail in other presentations as part of this symposium. I want to emphasize that this is a partial list primarily of scientific and engineering applications. A similar list could easily be generated for operations research, economics, law, and so forth. Some of the areas are outside strict definitions of the fields of chemistry and chemical engineering, but I have included them to illustrate the breadth of potential applications in related disciplines. 

This particular demonstration module only incorporates decisions involving analysis of volatile and semivolatile organic compounds from water. These compounds are, by definition, volatile enough to be separated by gas chromatography (GC). The complete expert system will incorporate decisions based upon any type of chemical in any type of matrix and will also be capable of providing advice specifically for selected EPA methods commonly in use, i.e., EPA Methods 624, 625, 1624, 1625, the various non-mass spectrometric 600 Methods, etc. (Figure 1). 

At least in the power industry, the terms "monitoring" and "diagnostics" are often used interchangeably or without careful definition. Much confusion can arise when these terms are used. For purposes of this paper, these terms and the terms "expert system" and "malfunction" will be defined here. 

Expert systems, and artificial intelligence in general, are new fields whose breadth of application, and indeed, whose exact definitions, are not yet completely settled. It is sometimes claimed that no two experts on artificial intelligence agree exactly on what its definition is. Definitions of expert systems at least agree on the necessity for expertise, but even here there are differences in emphasis and in priority. 

The Definite Clause Grammar (DCG) formalism [7] is utilized throughout this project. Grammar rules are used in the expert system rules to recognize the general class of the parent molecule in the disconnection (c.y., cyclohexene). The class determines the patterns used to construct the resultant synthons (discussed in Section 4). 

In other ways expert systems programming is much different. Traditional principles of specification and organization are tested, in part, because the program undergoes evolutionary and sometimes revolutionary revisions as an understanding of the problem domain grows. Despite early detailed specification, the tendency of the specification and the project to evolve toward its final definition seems to be unavoidable. 

EXMAT - A Linked Network of Expert Systems for Materials Analysis. Seven individual expert systems comprise EXMAT (1) problem definition and analytical strategy (2) instrumental configuration and conditions (3) data generation (4) chemometric/search algorithms (5) results (6) interpretation (7) analytical goals. Dynamic headspace (DHS)/GC and pyrolysis GC (PGC)/concentrators... 

Chemical metabolism can be described qualitatively or quantitatively. Many scientists can make qualitative predictions of the likely excretion products or blood plasma metabolites in mammals, or a particular animal including man, based on accumulated knowledge and experience. Such knowledge, in its raw form, generally consists of structure-metabolism relationships that are frequently expressible as qualitative structure-based rules that may be encoded into computer-based expert systems (see Chapter 9 for a full definition). Examples of such systems, in their more fully developed commercial forms, are discussed toward the end of this chapter. 

According to Klaessens et al. [203] a program must be able to solve substantial problems in order to deserve the qualification expert system . Although the substantial problems are definitely there to be solved, and hence expert systems may be of use in chromatography, there is no program good enough yet to deserve the name. 

The programs described in this section mimic the human expert by seeking out patterns in data. They perhaps barely conform to the definition of an expert system as one that uses a generalized store of knowledge, but they are widely thought of as members of the class. The programs are included here partly for that reason and partly because they are important for their contributions to this area of research the output from their analysis modules can support human experts working on the development of rules for the systems described above. 

Chapter 9 closes with a generic definition and a critical assessment of expert systems and an outlook. 

The Initiation Phase for an expert system covers the tasks involved in problem definition and in determining the need for an automated solution. It explores characteristics of a problem that suggest an expert system solution. However, the conclusions drawn from this phase are usually written independent of any particular technology. 

The objectives for the Concept Phase include problem definition, requirements and feasibility. These objectives define the approach to solve the information processing problem. The first objective of the Concept Phase is to confirm the existence of the information processing or knowledge-intensive problem. The second objective is to identify high level requirements for a solution to the problem. These requirements should focus on the nature of the problem and the user s needs. The third objective is to determine the feasibility of an expert system solution to the problem. This requires a study of the applicability of expert systems to the project and the capabilities of other information technologies in comparison to the choice of an expert system. 

Validation techniques are the methods used to determine that the expert system conforms to the functional requirements and can be used as intended. The validation techniques should be identified in the Definition and Design Phase. Issues such as the need for external experts, and types and location of test cases should be thought out. 

The nature of expert systems and their development process presents particular difficulties for the application of the traditional method of validating software against comprehensive pre-specifications, since the final level of operation of an expert system is difficult to pre-determine. Testing by measuring the degree of comparability of the system s performance to the performance of human domain experts, by using a representative set of problem cases, is more practical and more in tune with the definition of expert systems. 

---