Showing papers in "Expert Systems With Applications in 1991"
TL;DR: The Fuzzy Cognitive Map (FCM) alternative to the expert system is presented and a method for combining credibility-weighted FCMs to achieve a single global knowledge base is shown.
Abstract: Complex social systems are difficult to represent. Relationships between social forces demand feedback. For example, the causal connection between commodity price and consumer demand is a feedback system. Price increase tends to decrease demand for some commodities. On the other hand, an increased demand tends to elevate price. A stable system settles into equilibrium. A dynamic system is needed to model shifts in equilibrium brought about by changes in the causal environment. The knowledge-based expert system lacks the intrinsic structure for modeling these effects. Internally, all expert systems depend on these representations. Some attempt to simulate unrestricted graphs with virtual registers and loop counters. The result is a semantic gap between the internal representation of the social system and the social system itself. This article presents the Fuzzy Cognitive Map (FCM) alternative to the expert system. We first describe the FCM. We then diagram several social systems. Finally, we show a method for combining credibility-weighted FCMs to achieve a single global knowledge base.
307 citations
TL;DR: The use of expert systems in rapidly pervading the field of accounting is discussed in this paper, where a taxonomy of the practice areas in which expert systems are now being used is presented.
Abstract: The use of expert systems in rapidly pervading the field of accounting. All of the “big six” international accounting firms are either using expert systems in their practices or have systems under development. This paper begins with a taxonomy of the practice areas in which expert systems are now being used. Accounting expert systems currently in use at the big six accounting firms are described. Future directions and concerns for expert systems development by public accounting firms and the accounting profession are discussed.
49 citations
TL;DR: The real-time expert system “Safety-Bag” is an integral part of the electronic railway interlocking system ELEKTRA, which contains a logic channel and a safety channel, which checks the commands according to safety rules.
Abstract: The real-time expert system “Safety-Bag” is an integral part of the electronic railway interlocking system ELEKTRA. The interlockinh system contains a logic channel, which processes commands, and a safety channel, which checks the commands according to safety rules. Commands which are only allowed under certain circumstances are carried out only if the instructions generated by processing the command in the logic channel are checked and committed by the safety channel. To minimize the possibility of common errors in both channels, different programming paradigms are used. The logic channel is implemented in a procedural programming language and the Safety-Bag is implemented in PAMELA, a rule-based expert system language suitable for real-time applications. The rule-oriented programming paradigm is very well suited to the Safety-Bag because the safety requirements themselves are represented by rules. The Safety-Bag runs on a standard ALCATEL process control computer.
38 citations
TL;DR: An approach to the specification of expert systems is proposed that is flexible, yet rigorous enough to cover the important features of a wide range of potential expert system applications.
Abstract: The success of numerous expert systems in practical applications warrants a more formal approach to their development and evaluation. Reliability assurance of expert systems requires a methodology for the specification and evaluation of these systems. Expert systems are a new class of software system, but some traditional techniques of software development may be adapted to their construction. However, the specification of an expert system differs from that of a more traditional software program in that parts of the specification are permitted to be only partially described when development starts. Specifications have two important purposes: as contracts between suppliers and clients, and as blueprints for implementation. A specification consists of a problem specification and a solution specification. The problem specification plays the role of contract and states explicitly what the problem to be solved is, and the constraints that the final product must satisfy. The solution specification plays the role of blueprint and has two major aspects: analyzing how a human expert solves the problem, and proposing an equivalent automated solution. We propose an approach to the specification of expert systems that is flexible, yet rigorous enough to cover the important features of a wide range of potential expert system applications. We describe fully each of the components of an expert system specification and we relate specification to the issues of evaluation and maintenance of expert systems.
37 citations
TL;DR: The fundamental architectures of artificial neural networks are outlined, an artificial neural network which has potentially broad use is details, and the results indicate that the neural network model performed well compared to an inductive rule-generating algorithm and logistic regression.
Abstract: With all the power of von Neumann serial computation, emulation of certain tasks performed routinely by humans has been painfully slow. Perhaps this is because there is evidence that human decision making results from massively parallel computation in the brain. This evidence has generated interest in constructing models of parallel computation which are called artificial neural networks. This paper outlines the fundamental architectures of artificial neural networks, details an artificial neural network which has potentially broad use, and reports on the results of its application to an important accounting problem that has been analyzed by other methods. The results indicate that the neural network model performed well compared to an inductive rule-generating algorithm and logistic regression.
35 citations
IBM1
TL;DR: The results of a survey whose purpose was to begin documenting some of the experiences and problems KBS developers have encountered suggest that current practices can be improved and some specific recommendations are included.
Abstract: The majority of the work in knowledge-based system verification and validation (KBS V&V) has focused on developing techniques and concepts for performing V&V on expert systems. Little information is available on what V&V practices are currently in use by expert system developers. This article describes the results of a survey whose purpose was to begin documenting some of the experiences and problems KBS developers have encountered. The results of the survey suggest that current practices can be improved and some specific recommendations are included. The recommendations include developing requirements for expert system V&V, addressing common issues, recommending a life cycle for expert systems development, addressing readability and modularity issues, and investigating applicability of analysis tools.
33 citations
TL;DR: XROUTE, a software system that demonstrates an integrated framework for synergistic combination of good mathematical models, knowledge-based systems, artificial neural networks, and adaptive genetic searches, in the domain of computer-aided vehicle routing and scheduling problems, is developed.
Abstract: The combination of good mathematical models, knowledge-based systems, artificial neural networks, and adaptive genetic searches are shown to be synergistic. Practical applications of this combination produces near-optimal results, which none of the individual methods can produce on its own. We have developed XROUTE, a software system that demonstrates an integrated framework for this synergism, in the domain of computer-aided vehicle routing and scheduling problems. The purpose of this system is to assist researchers and decision makers who are applying the mathematical models to a specific routing problem instance by “tuning” the models to the problem description. The neural network modules store knowledge of previously solved problems and their solutions which facilitates the process of arriving at solutions to new problems. The knowledge-based system stores partial solutions from various knowledge sources, like the neural network and genetic algorithm modules, in the working memory and closely supervises the solution process in heuristic mathematical models. XROUTE provides an experimental, exploratory framework that allows many variations, and compares the alternatives on problems with different characteristics. The resultant system is dynamic, expandable, and adaptive and typically outperforms alternative methods in computer-aided vehicle routing.
27 citations
TL;DR: An evaluation and selection methodology for expert system (ES) shells is illustrated that considers the possibility that, at any stage of the process, no expert system shell is suitable and that a system must be developed with programming languages such as LISP, PROLOG, or some conventional programming language.
Abstract: This article illustrates an evaluation and selection methodology for expert system (ES) shells. The methodology incorporates three stages: (1) ES shell screaning, (2) shell evaluation, and (3) assurance of final ES shell selection. Initially, developing a short list through screening of commercial shell products determines whether appropriate software exists and narrows the field of available expert system software products for detailed consideration. The second stage determines which of the remaining ES shells (the finalists) best meets the needs of the organization, from both functional and technical perspectives. The final stage compares user requirements with the features of the selected ES software by defining how these requirements will be satisfied by building expert system applications with the selected product. The methodology also considers the possibility that, at any stage of the process, no expert system shell is suitable and that a system must be developed with programming languages such as LISP, PROLOG, or some conventional programming language.
25 citations
TL;DR: Demons and objects are found to be useful devises to facilitate the organization, storage and application of intelligence for accounting database systems.
Abstract: The purpose of this paper is to survey and extend the use of Artificial Intelligence and expert systems in accounting database.. The paper elicits a number of concerns often voiced about accounting databases. The use of Artificial Intelligence and expert system is investigated as a basis to mitigate those problems. The literature is surveyed and extended. Demons and objects are found to be useful devises to facilitate the organization, storage and application of intelligence for accounting database systems. Models for their use are presented.
25 citations
20 citations
TL;DR: In a typical operating system, one-third of a program consists of branch instructions, which means a performance of a processor of a Typical operating system depends greatly on whether or not an instruction before and after a branch instruction can be executed in parallel.
Abstract: In a typical operating system, one-third of a program consists of branch instructions. This means a performance of a processor of a typical operating system depends greatly on whether or not an instruction before and after a branch instruction can be executed in parallel. In order to provide a high performance processor with parallel processing, provided is a structure with a plurality of operating units and a plurality of registers where a set of registers are specified with the same address. A selection sequence of registers is stored by a plurality of selection sequence storages. Contents of registers are determined or not depending on the information stored in a plurality of determination identification storages. A register is specified by a register selector according to the contents of the selection sequence storages. This register selector is also used to update the contents of the selection sequence storages. The contents of the determination identification storages are rewritten by a determination identifier when the contents of a register proves to be a correct result.
TL;DR: The motivations for acquiring the knowledge of multiple experts, the problems related to knowledge acquisition, new issues that arise whens multiple experts interact, solutions that can be brought to bear in building multiple expert systems (particularly when experts are geographically dispersed), and new tools for knowledge engineers to use when dealing with multiple experts are discussed.
Abstract: The increasing complecity of many expert system application areas calls for the integration of the knowledge of multiple experts. The use of multiple experts introduces some interesting new problems during the process of knowledge acquisition. The problems are further complicated when the experts are geographically dispersed or unavailable for face-to-face interactions. This article discusses the motivations for acquiring the knowledge of multiple experts, the problems related to knowledge acquisition, new issues that arise whens multiple experts interact, solutions that can be brought to bear in building multiple expert systems (particularly when experts are geographically dispersed), and new tools for knowledge engineers to use when dealing with multiple experts.
[...]
TL;DR: By providing significant new capabilities to support both the development and documentation of design solutions, Design ++ increases productivity in design projects by 10 to 15 times, freeing engineers to make fuller use of their skills and creativity.
Abstract: Since their introduction in the eighties, Computer-Aided Drafting (CAD) systems have become ubiquitous among design engineers. From costly systems with simple interactive geometric capabilities, CAD systems have evolved into affordable but sophisticated tools that automate many aspects of design documentation and retrieval. The development of even routine designs, however, has still been a manual (and often tedious) process. These first generation CAD tools have provided gains in drafting productivity of 50% to 200% in repetitive design applications. This article describes the history and evolution of a new kind of knowledge-based CAD system, Design ++ . By providing significant new capabilities to support both the development and documentation of design solutions, Design ++ increases productivity in design projects by 10 to 15 times, freeing engineers to make fuller use of their skills and creativity. Design ++ allows storage and reuse of engineering knowledge, supports engineering teamwork and concurrent engineering, enables what-if experimentation, integrates engineering disciplines, and automates generation of documents, including drawings, parts lists, and specifications.
TL;DR: A new decision support framework for a disaster management is presented and it is demonstrated how an emergency response system can be designed to test correctness of a disaster manager's actions.
Abstract: This article presents a new decision support framework for a disaster management. This framework is established by NEGOPLAN which is a computer implementation of a rule-based model of sequential decisionmaking. The power and flexibility of NEGOPLAN are illustrated with a simple example. It is demonstrated how an emergency response system can be designed to test correctness of a disaster manager's actions.
TL;DR: SIPDES is an expert system that has been developed to aid in the debugging of simulation models written in Pascal and are hand written or derived from the use of an interactive simulation program generator.
Abstract: SIPDES is an expert system that has been developed to aid in the debugging of simulation models. The simulation models are written in Pascal and are hand written or are derived from the use of an interactive simulation program generator. User amendments oo the program occur because either the generator can not handle all of the model complexity, or because of a revision to the problem description. These amendments give rise to run time errors or errors seen at the reporting stage of the program run. An expert system was considered to be a suitable method for developing a debugging tool because of the limited domain of the problem (simulation and Pascal); because of the versatility of these systems in handling changes and extensions; and because of the limited availability of experts who had to acquire their expertise empirically. Experiences of system design, knowledge acquisition, and system validation are related. Future uses for such systems are outlined. The type of expert system development tool used is also discussed.
TL;DR: Experiments with adaptive neural networks which explore two types of learning, deductive and inductive, in the context of a rule-based, deterministic parser of Natural Language, and draw some general conclusions that are relevant to knowledge engineering activities and maintenance of rule- based systems.
Abstract: Rule-based expert systems either develop out of the direct involvement of a concerned expert or through the enormous efforts of intermediaries called knowledge engineers. In either case, knowledge engineering tools are inadequate in many ways to support the complex problem of expert system building. This article describes a set of experiments with adaptive neural networks which explore two types of learning, deductive and inductive, in the context of a rule-based, deterministic parser of Natural Language. Rule-based processing of Language is an important and complex domain. Experiences gained in this domain generalize to other rule-based domains. We report on those experiences and draw some general conclusions that are relevant to knowledge engineering activities and maintenance of rule-based systems.
TL;DR: It is argued that developers of intelligent help desk support systems (IHDSS) must consider several challenges not typically encountered in other environments, and the approach taken in developing Expertech, a partial implementation of a model of human expertise in network diagnosis.
Abstract: As more companies turn to expert systems to support traditional help desk activities, the need to understand the particular issues inherent in the development of such systems becomes increasingly important. This article describes the design and development of Expertech, a prototype expert system intended to assist a help desk operator in the resolution of common problems encountered by the users of a telecommunications network. Expertech, a rule-based system, is a partial implementation of a model of human expertise in network diagnosis. We argue that developers of intelligent help desk support systems (IHDSS) must consider several challenges not typically encountered in other environments. An IHDSS, for example, must consider the heterogeneous nature of the end users, and tailor its interface to encompass the varying technical knowledge of those users. Furthermore, an IHDSS must deal with the three-party nature of computer-based support, where the system does not interact directly with the end user. Instead, communication between the IHDSS and the end user takes place indirectly, through a human intermediary, the help desk operator. In the context of these considerations, we describe the approach we took in developing Expertech.
TL;DR: This work presents a distributed expert telecommunications provisioning system which uses a simulation-based optimization methodology for queueing networks and finds a random search based on the relative order of the performance gradient components to be a powerful qualitative reasoning technique.
Abstract: The area of telecommunications network design and management is complex and solutions can become algorithmically intractable for moderately large networks. It is, therefore, a promising applications area for expert systems; however, a survey of the published literature reveals a paucity of integrated systems combining design and optimization of network-based problems. We present a distributed expert telecommunications provisioning system which uses a simulation-based optimization methodology for queueing networks. Our architecture admits parallel simulation of multiple configurations. A knowledge-based search drives our performance optimization of the network. The search process is a randomized combination of Steepest Descent and Branch and Bound algorithms, where the generating function of new states uses qualitative reasoning, and the gradient of the objective function is estimated using a heuristic Score Function method. We found a random search based on the relative order of the performance gradient components to be a powerful qualitative reasoning technique. The system (P3) is implemented as a loosely coupled expert system with components written in PROLOG, SIMSCRIPT, and C. We demonstrate the efficacy of our method through an example from the domain of Jackson queueing networks.
TL;DR: The paper provides a basis for the standardization of the validation of knowledge-based systems using a model for validation of KBS causal processes as well as performance outcomes.
Abstract: The paper provides a basis for the standardization of the validation of knowledge-based systems. The place of validation within the development process is discussed and a model is proposed. Knowledge-based system validation problems are decomposed using sequences of independent and dependent generic tasks. A model for validation of KBS causal processes as well as performance outcomes is presented. Practical applications to real systems are described.
TL;DR: The problem of southern pine beetle management is defined, the history of ISPBEX development is recounted, the structure and function of the implemented version of ISPsBEX are examined, and the knowledge system environment approach is described that is being used in development of the next version of the system.
Abstract: The Integrated Southern Pine Beetle Expert System (ISPBEX) was developed to address the fundamental issue of pest management: how to use existing and forthcoming information for the purposes of problem solving and decision making. In this article, we define the problem of southern pine beetle management, recount the history of ISPBEX development, examine the structure and function of the implemented version of ISPBEX, and describe the knowledge system environment approach that is being used in development of the next version of the system. ISPBEX is a task-oriented knowledge-based system. It considers three aspects of integrated pest management : suppression, prevention, and utilization. These subject areas are represented as modules in ISPBEX. Each module contains specific advisors. Major emphasis has been placed on the suppression module which consists of five advisors that deal with treatments, presuppression economic evaluation, postsuppression economic evaluation, infestation forecasting, and project budgeting. The treatment advisor was Beta tested during 1989. A new version of ISPBEX is currently under development. It is an implementation of the Knowledge System Environment (KSE) concept. This new system will provide for knowledge-based interpretation of each of the different types of information and will include a facility for analysis and display of spatially referenced data.
TL;DR: A methodology for analyzing the contents of a Prolog-type knowledge base and eliminating inconsistent and redundant logical elements, and two alternatives, Boolean algebra and the Karnaugh map, are presented as tools to identify and to examine redundancy.
Abstract: Because of the incremental and piecemeal nature of its construction, logical inconsistency and redundancy can be built inadvertently into a knowledge base This paper discusses a methodology for analyzing the contents of a PROLOG-type knowledge base and for eliminating inconsistent and redundant logical elements It introduces a graphical representation, the goal-fact network, of the logic required to infer a goal and describes the identification of inconsistency in that network Three increasingly general alternatives, Boolean algebra, the Karnaugh map, and the Quine-McCluskey algorithm, are presented as tools to identify and to eliminate redundancy
TL;DR: Artificial Intelligence technology, specifically expert systems, is described as a logical complement to the present mechanical decision model approaches on capital budgeting domains.
Abstract: Research on capital budgeting, or capital planning, for governments has focused on developing increasingly sophisticated decision models. However, broad-based acceptance of sophisticated decision models has been limited, at best. This paper described Artificial Intelligence technology, specifically expert systems, as a logical complement to the present mechanical decision model approaches. The unique attributes of government capital budgeting domains are presented as a construct for discussing the feasibility of adoption and implementation of expert systems. Because of the domain-specific nature of feasibility studies, no declarative conclusions are formed. Rather, nontechnical and objective assessments of the relevant decision variables are presented and supported by extensive references.
TL;DR: An artificial intelligence-based decision components, the “trusted assistant”, is proposed as a means of achieving a more effective intelligence balance within the system and it is shown that in many instances this trusted assistant can play a distinctly novel and vital role in the decision making process.
Abstract: In this article, the problem of decision making in a prototypical complex decision environment is cast into the conceptual framework of large-scale systemstheory. The problem of the lead decision maker within a decision team is described and an operational model is given. It is pointed out that this primarily real intelligence-only decision team is both knowlegable and adaptive, but not as reliable as might be required under conditions of high cognitive loading. An artificial intelligence-based decision components, the “trusted assistant”,is the proposed as a means of achieving a more effective intelligence balance within the system. The primarily structural features of this personal expert system are presented and considerations involved in incorporating it into present-day large-scale decision systems are discussed. It is shown that in many instances this trusted assistant can play a distinctly novel and vital role in the decision making process.
TL;DR: It is argued that in many areas guidelines are more appropriate than standards for user interface design.
Abstract: In recent years there has been a growing debate over whether or not standards should be produced for user system interfaces. Those in favor of standardization argue that standards in this area will result in more usable systems, while those against argue that standardization is neither practical nor desirable. The present paper reviews both sides of this debate in relation to expert systems. It argues that in many areas guidelines are more appropriate than standards for user interface design.
TL;DR: The methodology applied in this paper on the evaluation relates the life-cycle model of the development to the phases of evaluation and emphasises the importance of feedback from expert collaborators and potential end-users.
Abstract: The motivation for the development of the thyroid system has been to help clinicians manage the problem of information overload by supporting the utilization and interpretation of laboratory data. The output from the system is a report that includes a proposal for diagnosis and/or recommendations for extra tests. The diagnosis is proposed on the axis hypothyroidism-euthyroidism-hyperthyroidism. The motivation to evaluate this prototype system is to approach the specifications of the next generation system for the test interpretation in the domain of clinical chemistry. The methodology applied in this paper on the evaluation relates the life-cycle model of the development to the phases of evaluation. According to this methodology evaluation should be done through the whole development cycle. The life-cycle view to evaluation comprises a series of iterative phases during the design and development process and it emphasises the importance of feedback from expert collaborators and potential end-users. The evaluation study described in this paper describes development of a prototype system for thyroid disorders. Of special interest are the validity of the system in application and the acceptability of the system by the users.
TL;DR: This article describes an application, ARCHES (American Red Cross Health Education System), which was developed by the author and is currently in use at several Red Cross chapters in the New England region.
Abstract: The past few years have witnessed an increasing interest in both academia and industry to integrate database management systems and expert systems technologies to develop more intelligent applications. Conventional fourth-generation languages coupled with relational databases provide an attractive environment for designing and implementing database-driven applications that incorporate expert system techniques. This article describes such an application, ARCHES (American Red Cross Health Education System). This application was developed by the author and is currently in use at several Red Cross chapters in the New England region.
TL;DR: It is shown how DEVS-Ada can exploit the parallelism intrinsic in the multiple execution of simultaneous experiments required for model validation, which reduces the bottleneck that validation imposes in the model development process and enables greater confidence to be achieved in the results.
Abstract: This paper reviews the concepts of a theory of modelling and simulation that relate to the validation enterprise. The theory provides a vocabulary, concepts, and mathematically rigorous tools with which to tackle problems in simulation model validation. We have implemented a hierarchical, distributed, object-oriented, and knowledge-based modelling and simulation environment in Ada. The environment, DEVS-Ada, provides portability, a standard model specification language, the means to manage a model repository, the ability to reuse models, and distributed simulation. We show how DEVS-Ada can exploit the parallelism intrinsic in the multiple execution of simultaneous experiments required for model validation. Faster execution of such a computationally intensive process reduces the bottleneck that validation imposes in the model development process and enables greater confidence to be achieved in the results. We also discuss the desirability of a more global view of validation which requires parallel symbolic analysis of a newly created model relative to existing models in a model base.
TL;DR: The three major modules that currently comprise the Kineticist's Workbench are described: a numerical simulation module; a graphical analysis module that uses mechanism structure to make predictions about behavior; and a qualitative analysis modules that interprets the numerical results of simulation and produces qualitative descriptions of those results.
Abstract: The Kineticist's Workbench is a program combining symbolic and numerical techniques, and intended to assist chemists in simulating, understanding, and analyzing chemical reaction mechanisms. This paper describes the three major modules that currently comprise the Workbench: a numerical simulation module; a graphical analysis module that uses mechanism structure to make predictions about behavior; and a qualitative analysis module that interprets the numerical results of simulation and produces qualitative descriptions of those results.
TL;DR: A computer-aided software engineering method using artificial intelligence techniques to instantiate a concrete program from a given abstract expert system architecture, called expert system reification, which is suitable for use in a large range of problem-solving behaviors.
Abstract: The problem of automatically reprogramming an expert system either to adjust for solving different types of computational problems or to improve run-time performance is addressed. A computer-aided software engineering (CASE) method using artificial intelligence techniques to instantiate a concrete program from a given abstract expert system architecture is proposed. Our method, called expert system reification, is suitable for use in a large range of problem-solving behaviors. Reification combines meta-level programming with partial evaluation of the program. Expert system reification extends the above blend with a knowledge-based translation module, thus obtaining an efficient and portable expert system version. We demonstrate the method by applying it to STAREX, an electronic circuit pack troubleshooting expert system developed in PROLOG to derive automatically the corresponding C code, which is currently installed at one of the AT&T manufacturing facilities.