Data-directed expert system

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 ... 

In working through process control examples, we found that many calculations, data checks, rate checks and other computationally intensive tasks are done at the first level of inference. Considerations of computational efficiency led to a design utilizing two parallel processors with a shared memory (Figure 1). One of the processors is a 68010 programmed in C code. This processor performs computationally intensive, low level tasks which are directed by the expert system in the LISP processor. 

The evaluation and interpretation of NMR parameters to establish molecular structures is usually a tedious task. An alternative way to elucidate a molecular structure is to directly compare its measured NMR spectrum - serving here as a fingerprint of the investigated molecule - with the corresponding spectra of known compounds. An expert system combining a comprehensive data base of NMR spectra with associated structures, NMR spectra prediction and structure generators not only facilitates this part of the NMR analysis but makes structure elucidation more reliable and efficient. 

The software now uses structurally intrinsic parameters for only one QSAR model (LSER) and the results are used to predict one property (acute toxicity) to four aquatic species by one mechanism (nonreactive, non-polar narcosis) however, we intend to continue to refine our equations as databases grow, incorporate other models, predict other properties, and include other organisms. We will attempt to differentiate between modes of toxic action and improve our estimates accordingly. For the widely divergent classes of chemicals and types of environmental behavior, no one model will best describe every situation and no one species is the optimal organism to monitor. As the software evolves, the expert system should choose the best model based on the contaminant, the species, and the property to be predicted (e.g., toxicity or bioaccumulation). In addition, we envision an interactive screen system for data entry that will bypass the SMILES notation and allow the user to describe the molecule by posing a series of questions about the compound s backbone and functional groups. The responses will translate directly into values of LSER variables. 

The escape rate coefllcients of the iodine and noble gas isotopes identified in one PWR or BWR plant can be directly applied to other plants of the same type, provided that the essential conditions, for example, fuel rod linear heat ratings, are comparable or can be corrected for. By this means it became possible to evaluate the number of failed fuel rods in the core of an operating reactor on the basis of the fission product activity concentrations measured in the coolant early attempts in this area were reported by Schuster et al. (1977). Although these estimates were based only on empirical data, they permitted a rather trustworthy prediction, as can be seen from Fig. 4.6., where the predicted numbers of failed fuel rods are compared with those detected in the course of post-cycle examinations. These techniques have been considerably improved, on the basis of experimental results as well as of model development and calculations. Because of the great number of parameters influencing the escape of fission products from defective fuel rods, evaluation of the number and type of defects from the measured activity concentration of fission products in the primary coolant is a difficult task which can be performed reliably only by specialists with considerable experience in this field. Attempts were, therefore, imdertaken to develop computerized expert systems that could be applied routinely. Lewis et al. (1992) described the development of such a system by means of which information can be obtained on the number of defects and their... 

In an industrial environment, such as the automotive industry, the principal requirement is for rapid analysis of difficult samples by direct means i.e. with a minimum of sample preparation), followed by an informative analysis of the data without the need for extensive spectral interpretation. Quite apart from the obvious need for rapid sampling in such a regime, detailed individual spectral analysis cannot be undertaken, and must be replaced with a quantitative semi-expert system approach. The availability of such software systems for... 

Figure )-11 gives an ovemew of the building hloc ks of the ELECTRAS system ELECTRAS was designed for two levels of user experience. The novice part offers a guided data analysis for inexperienced users. Experienced users can analyze their data fast and directly using the expert mode. 

We will discuss the main features of the offered system. At the lower level the state of all potentially dangerous sites - both natural and technogenic will be controlled. The periodicity and the facilities for carrying out the control over these sites is the concern of special expert subsystems, established on the basis of technical documents (for technical sites) or of data from the Ministry for ecological protection (for natural objects). For example, the sites of the first category should be equipped with special checking for the reliability of the security systems located directly on the site and continuous supervision of the status. 

---