Expert system parts

Expert Systems Part 2, Dessy, R. E., Ed. Anal. Chem. 56, 1984, 1312A... 

The recognition ratios achieved by CBR systems developed as part of this project could not be bettered by either neural-network classifiers or rule-based expert system classifiers. In addition, CBR systems should be mote reliable than simple classifiers as they are programmed to recognise unknown data. The knowledge acquisition necessary to build CBR systems is less expensive than for expert systems, because it is simpler to describe the knowledge how to distinguish between certain types of data than to describe the whole data contents. 

At the start of the development, it had been intended use an expert system shell to implement this tool, however, after careful consideration, it was concluded that this was not the optimum strategy. An examination procedure can be considered as consisting of two parts fixed documentary information and variable parameters. For the fixed documentary information, a hypertext-like browser can be incorporated to provide point-and-click navigation through the standard. For the variable parameters, such as probe scanning paths, the decisions involved are too complex to be easily specified in a set of rules. Therefore a software module was developed to perfonn calculations on 3D geometric models, created fi om templates scaled by the user. 

Today, the use of CHIRBASE as a tool in aiding the chemist in the identification of appropriate CSPs has produced impressive and valuable results. Although recent developments diminish the need for domain expertise, today the user must possess a certain level of knowledge of analytical chemistry and chiral chromatography. Nevertheless, further refinements will notably reduce this required level of expertise. Part of this effort will include the design of an expert system which will provide rule sets for each CSP in a given sample search context. The expert system will also be able to query the user about the specific requisites for each sample (scale, solubility, etc.) and generate rules which will indicate a ranked list of CSPs as well their most suitable experimental conditions (mobile phase, temperature, pH, etc.). 

The computer has become an accepted part of our daily lives. Computer applications in applied polymer science now are focussing on modelling, simulation, robotics, and expert systems rather than on the traditional subject of laboratory instrument automation and data reduction. The availability of inexpensive computing power and of package software for many applications has allowed the scientist to develop sophisticated applications in many areas without the need for extensive program development. 

J. Klaessens, J. Sanders, B. Vandeginste and G. Kateman, LABGEN, expert system for knowledge based modelling of analytical laboratories. Part 2, Application to a laboratory for quality control. Anal. Chim. Acta, 222 (1989) 19-34. 

The most popular representation scheme in expert systems is the mle-based scheme. In a rule-based system the knowledge consists of a number of variables, also called attributes, to which a number of possible values are assigned. The rules are the functions that relate the different attributes with each other. A mle base consists of a number of If... Then... mles. The IF part contains the conditions that must be satisfied for the actions or conclusions in the THEN part to be valid. As an example, suppose we want to express in a mle that a compound is unstable in an alkaline solution if it contains an ester function. In a semi-formal way the mle can be written ... 

Scott DR (1995) Empirical pattern recognition/expert system approach for classification and identification of toxic organic compounds from low resolution mass spectra. In Chemometrics in environmental chemistry - applications. Vol 2, part H (Vol ed J Einax), Springer, Berlin Heidelberg New York, p 25... 

In most expert systems, only small sections of the information in the knowledge base are causal. Causal knowledge consists of statements of facts and relationships, linked to information that explains why these facts are true. The "why" part is often not required for the ES to function effectively and, therefore, not included in a working ES. 

Taken together, all of these points suggest that it might be possible to prepare a toolkit consisting of the essential components of an ES, apart from the knowledge base, and then fill it with application-specific data. This is such a useful way to work that in all expert systems there is a clean separation between the information that the ES manipulates and the tools that are required to perform that manipulation. This division between the part of an ES that changes between applications and that which is constant has led to the development of the expert system shell. 

Analytical chemistry in the new millennium will continue to develop greater degrees of sophistication. The use of automation, especially involving robots, for routine work will increase and the role of ever more powerful computers and software, such as intelligent expert systems, will be a dominant factor. Extreme miniaturisation of techniques (the analytical laboratory on a chip ) and sensors designed for specific tasks will make a big impact. Despite such advances, the importance of, and the need for, trained analytical chemists is set to continue into the foreseeable future and it is vital that universities and colleges play a full part in the provision of relevant courses of study. 

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. 

RuleMaster expert systems are represented as Radial programs. To build an expert system, domain knowledge is normally entered in two parts a module structure and the bodies of the modules. The structure defines the hierarchical organization of decisions used to solve the problem. The code within each module defines the details of one of these decisions. 

The prototype of QualAId currently in existence is one small part of the total framework needed for a useful expert system. The objective of QualAId is to provide advice on how much and what type of QA/QC is needed for various types of environmental analyses. The rules for determining these needs have been derived from the American Chemical Society (ACS) publication, "Principles of Environmental Analysis, (2) and from various protocols and recommendations of the U.S. Environmental Protection Agency (EPA). 

RuleMaker, a subsystem of RuleMaster, induces rules for all situations from examples that may cover only some of the cases. At the heart of the induction process is the creation of an induction file, which in part includes examples indicating what the expert system should do under different circumstances. Now, in the example above, THE RULES FOR CORRELATING VARIOUS CHEMICAL AND PHYSICAL PARAMETERS OF THE HAZARDOUS CHEMICALS TESTED WITH THE PROTECTIVE ABILITY OF THE SELECTED GLOVE MATERIALS ARE NOT KNOWN — THEY WILL HAVE TO BE INDUCED FR04 THE ANALYTICAL DATA. 

Diagnosis is accomplished by the expert system. The central part of the expert system is the rule base. The rule base consists of ideas, called nodes, and rules which interconnect them as shown in Figure 2. The upper node is the evidence the lower node is the conclusion. The rule between them will state that if the evidence is known to be true with absolute certainty, then the conclusion will be known to be true (or false) with a specific confidence. 

An expert system has been written which helps the agricultural chemist develop formulations for new biologically active chemicals. The decision making process is segmented into two parts. The first is which type of formulation to use. The second is how to make a formulation of that tyrpe with the chemical of interest. The knowledge base currently contains rules to determine which formulation type to try and how to make an emulsifiable concentrate. The next phase will add rules on how to make other types of formulations. The program also interfaces to several FORTRAN programs which perform calculations such as solubilities. 

Multilevel expert systems offer additional advantages over traditional expert systems. Multilevel expert systems draw on computational computer programs to solve parts of the problem. The Ag formulation expert system does this in the areas of computational chemistry, bookkeeping, and communication. 

---