A risk-based expert system development methodology has been developed at NASA Goddard Space Flight Center to provide guidance to project managers and technical personnel and to serve as a standard for developing exper...
详细信息
A risk-based expert system development methodology has been developed at NASA Goddard Space Flight Center to provide guidance to project managers and technical personnel and to serve as a standard for developing expert systems. Concerns expressed by managers of expert system development projects are discussed. A description is given of the expert system life cycle and the risk-based expert system development methodology. Features of this methodology that address specific managerial concerns are reviewed.< >
A discussion is presented of an architecture called InFuse (intelligent Fusion) for the multiple sensor fusion problem. InFuse exploits the notion of combining expert systems and neural networks to capture the advanta...
详细信息
A discussion is presented of an architecture called InFuse (intelligent Fusion) for the multiple sensor fusion problem. InFuse exploits the notion of combining expert systems and neural networks to capture the advantages of both technologies. The application involves the extraction of natural terrain features from imagery provided by multiple sensors. In addition to the imagery, terrain knowledge in geographical databases needs to be integrated. The uniqueness of the approach lies in its ability to combine symbolic data of spatial databases and domain-specific knowledge about sensor behavior in expert systems with example-induced learning from neural networks to achieve a high classification rate. Results are presented to demonstrate the power of the approach.< >
The prototype application of a distributed problem-solving framework (DPSF) is described. The DPSF is made up of the coarse-grained multiagent systems defined as a set of computational agents that autonomously execute...
详细信息
The prototype application of a distributed problem-solving framework (DPSF) is described. The DPSF is made up of the coarse-grained multiagent systems defined as a set of computational agents that autonomously execute. cooperate, and coordinate each other to solve the network management problems. This loosely coupled heterogeneous system is connected by a common bus. In the proposed architectural components of the DPSF, the blackboard modulus consists of a data blackboard and a control blackboard, and knowledge sources (KSs) are composed of a local-planning KS, a metaplanning KS, a communication KS, a domain KS and a constraint KS. It is found that the DPSF offers advantages of performance, reliability, extensibility and the ability to handle incomplete data and knowledge.< >
Based on the development of PRXDM, an expert system for the fault diagnosis of the PRX/A electronic telephone exchange, a development life-cycle for expert system development is proposed. Seven phases are identified: ...
详细信息
Based on the development of PRXDM, an expert system for the fault diagnosis of the PRX/A electronic telephone exchange, a development life-cycle for expert system development is proposed. Seven phases are identified: domain definition, general design, detailed design, prototype development, integration, field testing, and operation. Provision is made for the extensive iteration and incrementation that is an essential part of all expert system development. A framework for testing. verifying, and validating expert systems is presented, and relevant metrics that could be collected in each phase of development are suggested.< >
The design of a system capable of handling ad hoc user responses to system-initiated questions is considered. An approach called the expectation-driven response understanding paradigm was developed. In this paradigm, ...
详细信息
The design of a system capable of handling ad hoc user responses to system-initiated questions is considered. An approach called the expectation-driven response understanding paradigm was developed. In this paradigm, it is assumed that each system-generated question is accompanied by an expectation of what the user should respond. The system attempts to link this expectation with the user's response in order to quantify its appropriateness in relation to the system's original question. A prototype system based on this paradigm has been implemented and integrated with the INGRES database system.< >
Signal decision models based on Dempster-Shafer and fuzzy set theories were developed to facilitate the implementation of a rule-based interpretation system for sensory evoked potential waveforms. The models do not re...
详细信息
Signal decision models based on Dempster-Shafer and fuzzy set theories were developed to facilitate the implementation of a rule-based interpretation system for sensory evoked potential waveforms. The models do not require a priori probabilities, and they provide measures of uncertainty. The two models were applied to a set of brainstem auditory evoked potential waveforms. Both had lower absolute error rates than discriminant analysis, although they yielded large numbers of uncertain waveforms. The fuzzy set approach may be less sensitive to the exact form of the membership functions.< >
A tool supporting a structured methodology for developing expert systems projects is presented. The tool is based on a set of forms, which represent the means by which the methodology can be implemented. Consequently,...
详细信息
A tool supporting a structured methodology for developing expert systems projects is presented. The tool is based on a set of forms, which represent the means by which the methodology can be implemented. Consequently, all the information acquired during the project can be gathered in a structured manner so that guidelines for system development and management can be obtained while filling them. Since some of them can be modified or redefined during the development of the project, multiple samples of the forms are collected, in order to record and track the decision-making process. The forms describe the whole development process, and also constitute the basic documentation for successive system maintenance.< >
State-of-the-art research and implementation efforts being conducted in the area of expert database systems with the goal of developingintelligent and effective applications is examined. Various approaches taken to i...
详细信息
ISBN:
(纸本)0818620595
State-of-the-art research and implementation efforts being conducted in the area of expert database systems with the goal of developingintelligent and effective applications is examined. Various approaches taken to integrate expert systems and database systems are discussed. Next, a real-world application developed by the author using one of the outlined techniques is discussed in detail. This application consists in the implementation of a comprehensive computerized system to automate the educational activities of the American Red Cross. The insights gleaned from this effort are presented. The most important lesson learned was to stop thinking about rules in abstract terms and visualize them instead, as any other valid entity (or object) as a part of the overall data model.
The author explores a general framework for analyzing or synthesizing intelligentsystems, as well as techniques for implementing them. The framework consists of six factors: purpose, space, structure, time, process a...
详细信息
ISBN:
(纸本)0818620595
The author explores a general framework for analyzing or synthesizing intelligentsystems, as well as techniques for implementing them. The framework consists of six factors: purpose, space, structure, time, process and efficiency. The technologies involve the methods of artificial intelligence and strategies for developing knowledge-based systems. The framework has been used to develop diverse applications, ranging from an expert system for creative product design to a package for factory configurations. This framework is discussed in the context of an expert system for the design of personal computers.
An approach to developing an analytically based methodology for effectively automating and supporting reasoning in environments characterized by uncertainty, imprecision, and vagueness is presented. The approach is ap...
详细信息
ISBN:
(纸本)0879425970
An approach to developing an analytically based methodology for effectively automating and supporting reasoning in environments characterized by uncertainty, imprecision, and vagueness is presented. The approach is applied to three cases. For one of the cases the procedure provides a reasoning system capable of completely distinguishing among alternatives (a total order is generated on the space of alternatives). However, it does not provide confidence statements on the selection. For another case the procedure provides a user with complete confidence in the set of alternatives selected, but might only provide a partial order on the set of alternatives. Procedures for providing intermediate values of confidence and distinguishability are needed in both cases. For the third case, methods that would represent important advances to existing capabilities for managing imprecision are needed. Six extensions of the procedures are presented.
暂无评论