Expert systems: rule types

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. 

The current implementation of the UltraLink uses a centralized set of terminologies, concepts, and rules, which may not correspond to the needs of every user. To further increase the flexibility of the expert system, we will implement a personalized version of the UltraLink. This should allow users to personalize the terminology, concepts, relationships, and rules used to identify typed entities and thereby create the UltraLinks best suited for their daily work. This will also enable us to design precustomized UltraLinks specifically tuned for chemists, biologists, and physicians. 

Foulkes et al. (1988) have approached the synthesis of operating procedures from a more empirical angle. They have extended the work of Rivas and Rudd (1974) for the synthesis of complex pump and valve sequencing operations, relying on the use of logical propositions (implemented as rule-based expert systems), which capture the various types of constraints imposed on the states of a processing system. 

The final aim is to construct a formalized representation of the decision process. Decision trees and structured system analysis are possibilities. Some types of expert systems can derive their own rules from examples. These are described in Chapters 18 and 33. 

The information to which the rule is applied might be extracted from the knowledge base, it might be provided by the user in response to questions from the ES, or it may be provided by combining the two. An expert system that uses rule-based reasoning is, quite reasonably, known as a rule-based system. This is the most widely used form of expert system in science, and it is on this type of system that this chapter concentrates. 

TOGA uses the built-in numerical capabilities of Radial to compute functions of concentration values, which are used extensively in the rules. The ratio of hydrogen to acetylene concentration in the corona rule is a simple example of this. User-defined con xDund data types are used to handle blocks of data as a single named structure. These features are invaluable in building practical expert systems, but are not available with all packages. 

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). 

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. 

II. TYPES OF EXPERT SYSTEMS A. Automated Rule-Induction (ARI) Systems 1. The Nature of ARI Systems... 

A second approach to the problem of difficult to obtain measurements is knowledge-based or model-based control. Knowledge-based systems attempt to use various types of knowledge of the biological process (rules etc.) to supplement traditional mathematical control approaches.16 Expert systems are one type of knowledge-based control. Model-based control systems use a model of the process as part of the control algorithm their reliability depends on the accuracy of the model. 

KnowledgeBase is apart of the expert systems memory that stores domain-specific knowledge in rules, frames, or other types of knowledge representations. 

Expert System for the Characterization of Rock Types (ESCORT) is an expert system based on Bayesian rules providing probabilities for the occurrence of rock types based on geochemical and nongeochemical data. 

The requirements for automatic interpretation of SOPs mentioned already leads us to another approach that is a general use for any steps in a laboratory workflow. If we look at the operator-entering data, we have to keep several critical sources of errors in mind typing errors, data type errors, formats errors, and data limit errors. One valuable solution based on expert system technology is a system that verities the data entered by the operator on the basis of rules. 

Recently, Diercksen and Hall (1) presented the OpenMol Program a proposal for an open, flexible and intelligent software system for performing quantum chemical computations. Central to their proposal was the observation that there is a close relationship between an abstract data type operation and a production rule in a rule-based expert system. The aim of this paper is to explore the establishment of a sound theoretical foundation for this relationship. 

The operation rules are, therefore, insufficient in themselves to produce a flexible expert system. They need to be augmented by type rules which express more general knowledge about the algebra of the data types. Seven such type rules are given in figure 4 and are described below ... 

Many times, the acceptance criteria also depend on the type of knowledge to be captured. Sometimes an expert-system approach is chosen as a convenience to incorporate "hard knowledge, such as inflexible rules involving regulatory requirements (Stunder and Hlinka 1989). In such cases, the expert system approach has no... 

We developed the first expert system that incorporates a working set of rules for a type of QSAR referred to as a linear solvation energy relationship or LSER (13-17) to predict LSER variable values from SMILES string formalism. The program also uses these LSER results and information about toxicity to predict acute toxicity to four representative organisms the fathead minnow (Pimephales promelas), the crustaceans Daphnia magna and Daphnia pulex. and Photobacterium phosphoreum. the luminescent agent in the Microtox test. 

Inference Engine. The inference engine is the central program which manipulates the rules and facts in the knowledge base to reach conclusions. The structure of the inference engine depends strongly upon the type of knowledge base which the expert system incor-... 

Data-directed expert systems begin with a list of the facts known to be true, and see what conclusions can be drawn from those facts. This type of expert system uses a forward-chaining mechanism. Each rule in the knowledge base is tested to see if all of its IF clauses are contained in the list of known facts. When such a rule is found, the system adds the THEN-clauses from the rule to the list of known facts. All the rules in the knowledge base are scanned repetitively until no new facts can be concluded. An example of using forwardchaining is illustrated by a structure elucidation problem based on an IR spectrum ... 

A rule-based system is clearly superior from a pedagogical viewpoint. Such a system would be more open-ended since it could use periodic relations to extrapolate to related reactions. A rule-based system also has the capability of being inverted to be able to explain its answers, a necessity if the expert system is to be developed into an intelligent tutoring system. However, a rule-based system is also expected to have limitations in the number of predictions it can make, since rules must be formulated to cover many different types of reactions. 

---