Abstract: A model of a real-time intrusion-detection expert system capable of detecting break-ins, penetrations, and other forms of computer abuse is described. The model is based on the hypothesis that security violations can be detected by monitoring a system's audit records for abnormal patterns of system usage. The model includes profiles for representing the behavior of subjects with respect to objects in terms of metrics and statistical models, and rules for acquiring knowledge about this behavior from audit records and for detecting anomalous behavior. The model is independent of any particular system, application environment, system vulnerability, or type of intrusion, thereby providing a framework for a general-purpose intrusion-detection expert system.

Abstract: The major limitations in building large software have always been (a) its brittleness when confronted by problems that were not foreseen by its builders, and (by the amount of manpower required. The recent history of expert systems, for example highlights how constricting the brittleness and knowledge acquisition bottlenecks are. Moreover, standard software methodology (e.g., working from a detailed "spec") has proven of little use in AI, a field which by definition tackles ill- structured problems. How can these bottlenecks be widened? Attractive, elegant answers have included machine learning, automatic programming, and natural language understanding. But decades of work on such systems have convinced us that each of these approaches has difficulty "scaling up" for want a substantial base of real world knowledge.

Abstract: Expert systems are emerging as a powerful tool for decision making. Integrating expert systems with decision support systems may enhance the quality and efficiency of both computerized systems. This article examines possible connections between the two technologies and discusses some issues related to their integration.

Abstract: This paper deals with an expert system which can estimate possible fault sections using information from protective relays and circuit breakers. This system is applicable to dispatching centers and can help dispatchers to judge emergency situations as the first step in restoration procedures. When some faults occur, the system makes inferences based on both knowledge about protection systems and information on the operating protective relays and tripped circuit breakers. The system can give possible answers even in the case of multiple faults and false operations of relays and circuit breakers. This expert system is written in Prolog.

Abstract: In this paper we look at the evolutionary development of Computer Assisted Instruction from the early days of ‘linear programs’ up to the use of ‘expert systems’ in education and training. We present the basic principles of Intelligent Tutoring Systems (ITS) which are capable of rich interaction with the student, which know how to teach, and who and what they are teaching. We point out the need for knowledge representation formalisms which can support ITS and present one such formalism (production systems). In the framework presented we describe systems developed for the teaching of modern languages, electronic trouble shooting and computer programming. Finally we point out the shortcomings of ITS and identify areas where a consensus of opinion does not exist. © 1986 Wiley Periodicals, Inc. (Masoud Yazdani: Masoud Yazdani studied Computer Science at the University of Essex and Artificial Intelligence at the University of Sussex. He joined Exeter University in 1980 as a lecturer in Computer Science. His research work is in two areas of computational models of creativity and intelligent tutoring systems. This latter area has led to implementation of various systems using artificial intelligence techniques to teach arithmetic and foreign languages. He is Chairman of Intellect Limited, a company specialising in applications of AI and to educate industry in what AI can offer it. He has acted as a consultant to various industrial companies including Acorn Computers, GTE Inc. and ITT Engineering Support Centre. He is the Committee Secretary for the Society for the Study of Artificial Intelligence and Simulation of Behaviour (AISB). His books include New Horizons in Educational Computing, Artificial Intelligence: Human Effects, both published by Ellis Horwood Ltd. and Artificial Intelligence: Principles and Applications published by Chapman and Hall.)

Abstract: Simulation and expert systems are remarkably similar. Both employ various representations to model some aspect of an uncertain world, with the model being formed as a piece of com puter software. This is then employed to aid decision making. Ideas about combining simulation and expert systems are presented, and a taxonomy is developed. It is concluded that the most fruitful areas of cross-fertilization are advice-giving ex pert systems that assist the simulation scientist and simulation user, new simulation tools built from knowledge-based tools, and intelligent front ends for simulation packages. Advice-giving systems will increasingly be part of simulation environments, rather than stand alone. They will be aimed primarily at the in experienced simulationist. An example of a system developed in this vein, which advises on experimentation with transaction flow models, is presented. Regarding the use of knowledge-based tools, induction as an aid to the development of a discrete model is considered.

Abstract: This paper details the design and implementation of ANGY, a rule-based expert system in the domain of medical image processing. Given a subtracted digital angiogram of the chest, ANGY identifies and isolates the coronary vessels, while ignoring any nonvessel structures which may have arisen from noise, variations in background contrast, imperfect subtraction, and irrelevent anatomical detail. The overall system is modularized into three stages: the preprocessing stage and the two stages embodied in the expert itself. In the preprocessing stage, low-level image processing routines written in C are used to create a segmented representation of the input image. These routines are applied sequentially. The expert system is rule-based and is written in OPS5 and LISP. It is separated into two stages: The low-level image processing stage embodies a domain-independent knowledge of segmentation, grouping, and shape analysis. Working with both edges and regions, it determines such relations as parallel and adjacent and attempts to refine the segmentation begun by the preprocessing. The high-level medical stage embodies a domain-dependent knowledge of cardiac anatomy and physiology. Applying this knowledge to the objects and relations determined in the preceding two stages, it identifies those objects which are vessels and eliminates all others.

Abstract: An expert system is developed to assist in the decision-making of the reactive power/voltage control problem. The knowledge required to perform the task is identified. To alleviate a voltage problem, empirical rules are used to generate appropriate control actions when slight voltage violations occur. Controls such as shunt capacitors, transformer tap changers and generator voltages are utilized. Theoretical justification of the empirical rules is presented. When it is judged that the voltage problem is so severe that empirical judgements may not be reliable, the developed expert system can help in formulating the problem so that an available application software package can be effectively utilized. In this paper, production rules are proposed to perform the above functions. Numerical examples are also presented.

Abstract: Techniques in developing a coherent probabilistic reasoning system are illustrated with reference to a simplified example. Recent work relating statistical models to graphical representation of causal and associative relationships allows a straightforward means of propagating evidence whilst retaining a probabilistic interpretation for predictive statements. This interpretation allows continual criticism of a system's performance, while imprecise quantitative assessments permit learning from experience. Possible limitations of a formal probabilistic approach are discussed.

Abstract: Expert systems are computer programs that perform at the level of human experts. One expert system, Comax, has been developed that acts as an expert in cotton crop management. The system has a knowledge base consisting of a sophisticated cotton plant simulation computer program, a set of "if-then" rules, and a computer program called an inference engine. Comax determines the best strategy for irrigating, applying fertilizer, and applying defoliants and cotton boll openers. Sensors in the cotton fields automatically report weather conditions to the system, and Comax reevaluates its recommendations daily. Comax was tested on a large farm and demonstrated excellent results in reducing the unit costs of production.

Abstract: Alarm processing has been a traditional feature of energy management systems and has not changed significantly over several generations of EMS design. Problems are obvious, however, and operations personnel have often voiced a desire for a better way to monitor a power system than provided by existing alarm processing software and hardware. The thesis of this paper and the experiments reported show promise of a truly different methodology for handling alarms. This is realized through the application of a realtime expert system dedicated to continuous analysis and reporting of system conditions rather than simply printing numerous specific alarm messages. Results show the ability of the expert system to present operators with concise alarm information extracted from a standard set of alarm messages.

Abstract: Many expert systems have moved out of development laboratories into field test and routine use. About sixty such systems are listed. Academic research laboratories are contributing manpower to fuel the commercial development of AI. But the quantity of AI research may decline as a result unless the applied systems are experimented with and analyzed.

Abstract: In this article, we present a programming language for expressing classificatory problem solvers. CSRL (Conceptual Structures Representation Language) provides structures for representing classification trees, for navigating within those trees, and for encoding uncertainly judgments about the presence of hypotheses. We discuss the motivations, theory, and assumptions that underlie CRSL. Also, some expert systems constructed with CSRL are briefly described.

Abstract: Microcomputers can host real-time control expert systems. As Hexscon shows, the key is combining the best of conventional and expert-systm controllers.

Abstract: Evaluation of expert systems is an important step in the knowledge engineering process. In recent years, emphasis has been placed on knowledge acquisition, knowledge representation, and inferencing mechanisms. However, evaluation, and to a lesser extent, validation, have been slightly overlooked. This paper addresses some of the evaluation techniques that have been used for measuring information systems effectiveness and expert systems effectiveness. Specific attention is focused on the Analytic Hierarchy Process and Expert Choice. The evaluation results of an expert system prototype, called ‘Read’, are discussed via the use of the Analytic Hierarchy Process and Expert Choice.

Abstract: This paper describes two knowledge-based programs. The first simulates the behavior of automatic protection schemes in power networks. The second is an expert system for the diagnosis of faults. Both are coded in OPS5--a widely available language for writing rule-based programs. The user and the programs communicate over a Blackboard which is a database for messages. The Blackboard has been organized so that the addition of new programs, whether knowledge-based or algorithmic, will be relatively easy.

Abstract: The use of knowledge engineering in diagnostic systems, is aimed primarily at exploiting procedural knowledge (about: systems operations, configuration, observations, calibration, maintenance), in connection with failure detection and test generation tasks. Next, the goal is to devise knowledge representation schemes whereby the failure events can be analyzed by merging highly diverse sources of information: analog/digital signals, logical variables and test outcomes, text from verbal reports, and inspection images. The final goal is to ease the operator workload when interfacing with the system under test and/or the test equipment, or with reliability assessment software packages. The paper will present key notions, methods and tools from: knowledge representation, inference procedures, pattern analysis. This will be illustrated by reference to a number of current and potential applications for e.g.: electronics failure detection, control systems testing, analysis of intermittent failures, false alarm reduction, test generation, maintenance trainers.

Abstract: In this article we introduce a formal procedure for the use of expert opinions in reliability (and fault tree) analysis. We consider the case of multicomponent parallel redundant systems for which there could be a single expert or a group of experts giving us opinions about each component. Inherent in our approach are a procedure for reflecting our judgment of the experts' expertise and our own opinions about the components' life lengths. An important issue here is the induced dependencies between the components' life lengths due to any common knowledge shared by the experts. Our final results are approximations that depend on our having small uncertainty about what the experts have to say. The approximations are easily computable, and they can be generalized to cover any coherent system.

Abstract: This two part paper explores management issues raised by expert systems (ES). In the first section, a brief history of ES is presented, and the competitive potential of ES is analyzed from a business policy perspective. The second section, appearing in the next issue, discusses development and implementation issues of ES. Both sections highlight similarities and differences between expert systems and other types of information systems.

Abstract: Many design problems can be formulated as a process of searching a "well-defined" space of artifacts with similar functionality. The dimensions of such spaces are largely known and are constrained by relations obtained from the implicit functionality of the designed artifact. After identifying the kinds of knowledge that mediate the search for acceptable designs, a computational framework is presented that organizes the required knowledge as design plans. A problem solver is described that executes these plans. The problem solver extends the notion of dependency-directed backtracking with an advice mechanism. This mechanism allows information from a constraint failure to be used as advice in modifying a partial design. An expert system for designing paper transports inside copiers has been successfully built based on this framework.