Explanation in expert system shells: a tool for exploration and learning

The use of an expert system shell (EXPERTECH Xi Plus) in the construction of an expert system for the diagnosis of infertility has been evaluated. A module was devised for predicting ovulation from the medical history alone. Two versions of this system were constructed, one using the expert system shell, and the other using QuickBASIC. The two systems have been compared with respect to: (1) ease of construction; (2) ease of knowledge base update; (3) help and explanation facilities; (4) diagnostic accuracy; (5) acceptability to patients and clinicians; (6) user-friendliness and ease of use; (7) use of memory space; and (8) run time. The responses of patients and clinicians were evaluated by questionnaires. The predictions made by the computer systems were compared to the conclusions reached by clinicians and to the "gold standard" of day 21 progesterone. The conclusions of this pilot study are: (1) the construction of this expert system was NOT facilitated by the use of this expert system shell; (2) update of the knowledge base was not facilitated either; (3) the expert system shell offered built-in help and explanation facilities, but as the system increased in complexity these became less useful; (4) after initial adjustment of decision thresholds the diagnostic accuracy of the system equalled that of the clinician; (5) the patient response to computer history-taking was very favorable but much less favorable to computer diagnosis; (6) the clinicians took a positive attitude to computer diagnosis; (7) the systems were easy to use; (8) the expert systems shell required much more memory space and had a much slower response time than the system written in BASIC.

Although the use of expert systems (ESs) as a problem-solving tool is relatively new in the transportation profession, it has been recognized that a well-designed ES can contribute in areas where knowledge domain requires an integration of expertise from various disciplines or is very complex and difficult to quantify. To minimize the effort involved in developing essential ES mechanisms, researchers often prefer to use ES shells which provide a standard inference procedure, knowledge representation, and a user-friendly design environment. However, commercial ES shells tend to be designed for only certain classes of problems, and are not as general and flexible as programming languages. More than 150 commercial ES shells are available in the microcomputer market. Most ES shells are ususally claimed by their vendors to be usable for almost any conceivable applications, but such claims have often proved to be extravagant. Hence, the selection of an efficient yet cost-effective commercial shell is one of the most critical tasks in ES development. In reviewing the the vital task of ES shell selection, this paper presents our experience in evaluating more than 30 ES shells for developing an ES for signal design. The proposed ES includes signal settings, phasing plans, controllers, and detector locations.

Expert system knowledge represents expertise obtained through formal education, training, and/or experience. Formal education provides deep knowledge of a particular domain; experience and training result in heuristic knowledge. A knowledge base defines the range of information and understanding with which the system is capable of dealing; therefore, its information must be structured and filed for ready access. The objective of this symposium is to address the challenges associated with establishment of valid expert system knowledge, specifically, knowledge to be used by expert system shells. As expert system knowledge is obtained, structured, and stored, it is formulated. In this symposium, knowledge formulation is addressed as a three-phase process: knowledge acquisition, the mechanics associated with structuring knowledge, and knowledge porting. Knowledge acquisition is the process of extracting expertise from a domain expert. Expertise may be collected through a series of interviews between the expert and a knowledge engineer or through sessions the expert holds with an automated knowledge acquisition tool. Thus, the ultimate outcome of knowledge acquisition is a collection of raw knowledge data. The following human factors issues become apparent: documenting mental models (where mental models are the expert's conceptualization of a problem), recording cognitive problem-solving strategies, and specifying an appropriate interface between the domain expert and the acquisition methodology. The knowledge structuring process involves the refinement of raw knowledge data, where knowledge is organized and assigned a semantic structure. One issue that must be considered is how to interpret knowledge data such that formal definitions, logical relationships, and facts can be established. Finally, formulation involves knowledge porting, that is, the movement of an expert system shell's knowledge base to various other shells. The outcome of this process is a portable knowledge base, where the challenges lie in maintaining consistent knowledge, understanding the constraints inherent to a shell (the shell's ability to incorporate all relevant knowledge), and designing an acceptable user-expert system interface. The fundamental component of any expert system is its knowledge base. The issues to be presented in this symposium are important because they address three processes that are critical to the development of a knowledge base. In addition to presenting computer science challenges, knowledge base formulation also presents human factors challenges, for example, understanding cognitive problem-solving processes, representing uncertain information, and defining human-expert system interface problems. This symposium will provide a forum for discussion of both types of challenges.

Expert-system technology has emerged from beneath the umbrella of Artificial Intelligence and into the mainstream of engineering problem solving. There already are several notable examples of the application of this state-of-the-art technology in the petroleum industry. Clearly, expert systems that run efficiently on standard personal computers are particularly of value. A new expert-system generator that overcomes the speed, memory and mass-storage limitations of XT/AT-class personal computers is presented in this paper. The shell is optimized for the specific characteristics of the target user, knowledge expert and computer. Expert systems developed using this shell currently are being used by our technical personnel for drilling fluid and related applications.

Expert systems have now proven their value in a variety of enterprises ranging from medicine to geology. Small expert systems are at work in numerous commercial and industrial sites, but large systems are still rare. Besides the difficulties inherent in the collection of useful and correct domain knowledge, large knowledge bases present special problems in data management. These problems often cause a good pilot system to founder when expanded to an operational size.This paper addresses the difficulties of efficiently handling large knowledge bases. The methods discussed result from viewing a knowledge base as a database and analyzing the knowledge as one would analyze a database application, using Chen's Entity-Relationship (E-R) Diagrams [CHEN76] to build a relational database representation. Besides providing a different perspective on the management of knowledge bases, such a method facilitates the use of a commercial DBMS for this task.Following these ideas the simple assertions of the knowledge base are viewed as one entity type and the rules themselves as another, with IF and THEN being separate relationships between these two types. This decomposition is reasonable for capturing the knowledge represented by the rules and leads to a relational description which is considerably different than usual rule representations. More precisely, the E-R diagram of Figure 1 is taken as the basis of the model, where the rectangles represent entities and the ellipses represent relationships between the entities connected to that ellipse. The unboxed words such as rule# are attributes of the associated entities and correspond to the information to be saved in the database.The E-R diagram in Figure 1 can be converted to the relational database tables shown in Figure 2 where rule# is a name or a number that identifies the rule, if_part is a simple assertion Ai (antecedent), then_part is a simple assertion Ci (conclusion), and conf is a number representing the confidence in that deduction.This results in a very different data organization from the usual collection of knowledge found in rule-based systems. Beyond that analysis, very specific implementation techniques are suggested for knowledge bases, using both standard relational data bases and simpler file-handling techniques such as those available in standard programming languages. Caution must be taken, however, before simply trying to turn knowledge base management over to a commercial DBMS, since the behavior of “queries” within a rule base is very different from that of ordinary database queries. Comparisons are made to such recent works as those of Deering, Debanne, and Bouzegoub which suggest a variety of implementations for knowledge bases.Often in the modeling of a domain, one discovers a hierarchical relation among the basic concepts, vocabulary, or objects of study. Such an IsA or AKindOf hierarchy can be used to create a segmentation of the knowledge base resulting in significant efficiency improvements within the system. This possibility is explored in some detail.Throughout this work, emphasis is given to dynamic rule bases which are subject to frequent changes and which, therefore, require a generalized inference engine and flexible data structures. Often due to execution speed requirements some domain knowledge is encoded as procedures within the inference engine. Although this may improve the efficiency of the system, it undermines the generality and flexibility that are also required. This paper investigates a promising method for incorporating such procedural inclusions within the general knowledge base structure. The approach developed allows the absorption of changes without restructuring of the knowledge base or modifications to the inference engine, and maintains system transparency, simplicity, and expandability.

An expert system for selecting the methods for retrofitting fatigue cracking in steel bridges is developed. The expert system involves a knowledge base with new knowledge representations and a new inference engine. The knowledge is based on 90 cases of fatigue cracking in existing steel bridges, and a knowledge base is constructed with knowledge being represented as production relations. All the relations in the knowledge base are assumed to have a certainty factor. The new inference engine was developed in the C language so that it has the ability to learn by using the knowledge‐based network model. The inference engine is able to modify relations with a certainty factor. Therefore, the knowledge‐based network model can grow into an optimum knowledge‐based network. The present system is evaluated for the actual fatigue cracking in three steel bridges. It is found that the system is able to infer reasonable methods for retrofitting cracked members.

Nowadays, expert systems being used in various fields have received a great deal of attention. Auditing is one such field, along with determining the audit opinion type. An expert system consists of a knowledge database and an inference engine. The objective of this research is to make an expert system that will be of help to auditors in predicting and determining the different types of audit reports. The expert system receives data or knowledge from financial reports and determines the types of audit opinions by using an artificial neural network and a decision tree as an inference engine. An expert system should be able to explain the solution, but presenting the reason for the results obtained with a neural network is difficult. This study attempts to provide a method that will present simple and understandable reasons for the results obtained with neural networks.

An integrated expert decision support system for agricultural management

This research paper presents a new approach to the diagnosis of dengue patients based on a new Mediative neuro-fuzzy method for Sugeno’s inference system. We design a Mediative Neuro-Fuzzy System (MNFS) to adapt the antecedent and consequent part of the inference fuzzy system. We also used a fuzzy optimization technique to optimize the coefficients of the functions which represents the consequent part used for Sugeno’s inference system. This article represents a combined study between Sugeno’s fuzzy expert system and Mediative based neuro-fuzzy system with fuzzy optimization technique for the diagnosis of dengue in the patients. We have also introduced a new concept for the formulation of membership/non-membership function of the input data on the basis of the trend of data by curve fitting. We applied our method on the input data set which has been collected from Lala Lajpat Rai Memorial Medical College (LLRMC) located in Meerut, Uttar Pradesh, India. These input factors include Temperature, Sugar, Pulse rate (PR), Age, Weight, Cough, Laboratory reports of Dengue, Chills, Headache, Blood Pressure (BP), Muscle pain and Chest pain. The output factors have also been classified into three attributes with the condition of the patients. We also include some numeric computation at the end of the research paper. The main objective of this research paper is to develop a mathematical model for the diagnosis of dengue patients based on Neural Network (NN) models.

Rule-based systems are relatively easier to explain to end users because rules can be expressed as if/else conditions. Indeed, at times there are multiple if/else conditions. Expert systems are deterministic in nature and very accurate, because they are driven by well-established rules and conditions. The need for explainability arises when multiple rules either conflict or are combined in such a way that it is difficult to interpret. In the field of artificial intelligence, an expert system is a program designed to simulate a real decision making process. Expert systems are designed to solve complex problems by simple reasoning and are represented as if/else/then conditions. Expert systems are designed to act like an inference engine. The role of an inference engine is to make decisions. Sometimes expert systems are simple, like a decision tree, and sometimes expert systems are very complex. Complex expert systems require a knowledge base. Sometimes these knowledge bases are an ontology, which is a form of a body of knowledge, through which a particular rule can be framed and inferencing can be done.

An expert database system is an expert system which is developed on databases using DBMS technology to manage facts and rules. Frame-based expert systems are widely used as the knowledge representation for such expert systems with large knowledge bases. Many systems have the ability to connect to external databases. Facts stored in databases can be loaded into expert system's knowledge base and inference is performed by the inference engine of the expert system. In many cases, such external facts are required several times for each inference. Thus, a lot of communication traffic takes place. This research work present the design and implementation of a frame-based object-relational database system which has a tight coupling between the expert system and the external knowledge base. The external knowledge base also use frame as its knowledge representation. More over, it has its own inference engine so that inference can be perform on the knowledge base side and the results, not only simple facts, are sent back to the expert system for further inference. A medical consultation system is used as an illustrated example an Oracle lOg object relational DBMS is used as the database platform.

This article is an introduction to artificial intelligence (AI), especially the place of expert systems (ES) in AI. Some applications of expert systems in other industries are discussed, as well as potential areas of application within the pharmaceutical industry. Advantages and disadvantages of developing ES using so-called AI languages v ES shells are covered. A product review is presented, including AI languages, induction ES shells, straightforward rule shells, and sophisticated ES development environments. Also included is a discussion of the problems and pitfalls that may await anyone attempting to develop ES in pharmaceutical R&D. Finally, some recommendations are offered that they may be useful to those just getting started in this area.

The application of an expert system (expert system) which is very useful for the wider community is certainly very much needed by society today. Apart from that, this development also supports various human activities. Expert systems are also able to make recommendations for a series of user actions. The aim of this research is to determine the development of expert systems in management information systems. The research method used in this research is a descriptive research method with theoretical studies, namely a method that discusses several possibilities for solving actual problems by collecting data, compiling or classifying them, analyzing and interpreting them with existing theories and comparing them from literature sources. The research results show the development of expert systems in management information systems, namely knowledge bases, inference engines, and user interfaces. It can be concluded that expert systems help work and technological progress more quickly and the decision-making process is based on existing information and facts. Expert systems have also been used in various fields, from medical, legal, to financial. In developing an expert system, it is necessary to increase the level of knowledge related to the latest technology and information. So you can utilize modern technology to complete work more effectively and efficiently.

An expert system is presented that produces the operation sequence for disconnecting apparatus from a power system under normal operating conditions.To maintain the reliability of power supply on the network during the operation, the operation sequence must be produced in due consideration of network states, such as load factors and routes of transmission line. Besides, since the network which the system manages includes many high‐voltage consumers, the operation sequence for each consumer should be divided into detailed switch operation sequence.To produce a detailed switching operation sequence simultaneously with macronetwork operation sequence, we introduce the multilayered object‐oriented network model for the knowledge base of this expert system.Each layer of this model represents network configuration in the different degree of detail, and the consistency between states of layers is maintained by an interlayer message‐passing mechanism. On the layer which represents apparatus and their connections in detail, the inference engine can generate the switching operation sequence. On another layer which represents control centers and consumers, the inference engine can search for routes of power supply and can generate the sequence of changing over the network. Therefore, by using each layer properly, the inference engine can produce various kinds of operation sequences.Furthermore the object‐oriented network model facilitates maintenance and extension of the system.

The Mission Operations Directorate (MOD) at the Johnson Space Center (JSC) is responsible for the safe operation and mission success of manned space flights. The MITRE Corporation, working with MOD, is developing requirements for the Operations Management System (OMS) to automate many aspects of flight control for the Space Station onboard systems. To help develop these requirements, the Integrated Status Assessment (ISA) expert system prototype was built to perform Station-wide failure diagnosis. This paper describes the OMS, how the ISA prototype was built, and how the prototype is being used to help define the OMS.Currently the Space Shuttle is managed on the ground using a hierarchy of control personnel. A flight director responsible for the overall Shuttle mission, a number of front room flight controllers are each responsible for a different system, and many back room controllers support various front room controllers. This system has been used since the early days of space flight and is very manpower intensive. A similar hierarchy is seen in an automated system for the Space Station. The back room controllers can be augmented with smart components and sensors on the Station and the front room controllers can have their jobs aided with expert systems. At the highest level is the OMS which works at the flight director level getting summary information from the various systems in the Station.The OMS will include onboard automation, ground based automation, onboard manual operations, and ground based operations. One of the functions of the OMS is to perform a Station-wide status assessment and failure diagnosis of the Station. This Station-wide view is very important because all of the systems interact with each other and a failure in one system will have impacts on the other systems.It became evident early in discussion with to the MOD flight controllers that prototypes would be very helpful in the process of gathering requirements for the OMS. The prototypes were used to gather additional requirements, to expose people to new ideas and technologies, and study the feasibility of these technologies for systems management.Because the task of assessing the status of space vehicles is a complex job that requires “expert” knowledge to find heuristic solutions to problems, an expert system approach was chosen to prototype the ISA system. The initial domain of the expert system was the KU band portion of the communications and tracking system (one of the onboard “core” Space Station Systems).The ISA prototype expert system consists of a knowledge base, an inference engine, and a user interface. The ISA system was designed as a hybrid expert system using different methodologies: object-oriented programming, rule-based programming using both shallow and deep reasoning, and qualitative modeling.The ISA prototype has been very successful in its goals. ISA has demonstrated new technology and ideas to the NASA operations community, helped define requirements for the OMS, and shown the feasibility of using hierarchical expert systems to monitor complex systems. The ISA prototype, along with other OMS prototypes, are being integrated in a test bed environment with other space station prototypes to form the Space Station Information System (SSIS) End-to-end Test Capability (ETC) which will have components all over the world. The results of the ETC will influence the Space Station design and be a test environment for software and hardware for years to come.