The Advanced Automation Methodology Project (AAMP), at NASA's Johnson Space Center, is described. The AAMP will develop two independent knowledge-based system applications, identify system features required to inc...
详细信息
The Advanced Automation Methodology Project (AAMP), at NASA's Johnson Space Center, is described. The AAMP will develop two independent knowledge-based system applications, identify system features required to incorporate artificial intelligence upgrades to several US Space Station Freedom systems in the future, and produce an engineering methodology for artificial intelligence system development in the Space Station Freedom program. The AAMP is composed of four teams: the management team, two development teams, and an independent test and verification team. By distributing project management responsibilities among the various management team members, the AAMP has realized several benefits. First, the diverse opinions and experience of the team often help focus management attention on the project's most critical areas. Second, this distributed approach allows team members to participate in the AAMP without having to dedicate full time to the project. Finally, by distributing management responsibilities, the AAMP has capitalized on each member's expertise.< >
The Log Quality Monitoring system (LQMS) is a medium-sized knowledge-based system that was developed to address the problem of providing real-time assistance to the wireline field engineer. The authors discuss the mot...
详细信息
The Log Quality Monitoring system (LQMS) is a medium-sized knowledge-based system that was developed to address the problem of providing real-time assistance to the wireline field engineer. The authors discuss the motivation for the project, review the major decisions that were made, detail how they affected the direction of the project, and identify the basic lessons that were learned. The research version of LQMS was developed at Schlumberger-Doll Research over a period of about three and a half years. The project successfully dealt with several thorny issues in the development of knowledge-based systems: multiple domain experts, real time constraints, embedded environment, coordination and communication of multiple knowledge sources, and technology transfer. The project evolved from an exploratory, rapid prototyping style of development to a more traditional specify/design/implement/test methodology. Much of this evolution was driven by the field testing and technology transfer requirements of the later stages of the project.< >
The importance of knowledge-based induction programs for problem solving is discussed. Desiderata for knowledge-based induction programs are given, and an example of such a program in the context learning classificati...
详细信息
ISBN:
(纸本)0818620846
The importance of knowledge-based induction programs for problem solving is discussed. Desiderata for knowledge-based induction programs are given, and an example of such a program in the context learning classifications is discussed. The induction program RL4 is used as an induction tool, and several examples of its past and present uses are presented. The power of the tool comes from its flexibility and ease of use with a performance system. The use of RL4 with an inference engine that uses user-defined or default evidence gathering strategies is also discussed. Finally, the directions in which RL4 can go in the future are considered.
Since the diagnosis of carcinoma in the esophagus routinely leads to surgery, the development of expertsystems which reduce diagnostic uncertainty is useful. In the present work, the histological features of esophage...
详细信息
ISBN:
(纸本)0879425598
Since the diagnosis of carcinoma in the esophagus routinely leads to surgery, the development of expertsystems which reduce diagnostic uncertainty is useful. In the present work, the histological features of esophageal tissue are quantified with microscopic image analysis (MIA), which facilitates statistical analysis and expertsystem development. Statistical discriminant analysis of MIA data is used to produce a model which delineates high or low survival in patients with esophageal adenocarcinoma. Based on this model, rule-based expertsystems, application programs, and example-based neural networks are compared. Each provides a useful risk assessment for patient survival after surgery on a training-test case set of esophageal adenocarcinoma.
A set of benchmark programs that have many of the characteristics of full-sized, complex knowledge-based systems are presented. They are: Weaver, a VLSI routing program;Rubik, which solves the Rubik's cube problem...
详细信息
ISBN:
(纸本)0818620447
A set of benchmark programs that have many of the characteristics of full-sized, complex knowledge-based systems are presented. They are: Weaver, a VLSI routing program;Rubik, which solves the Rubik's cube problem;Tourney, which assigns match schedules for a tournament;and Big Conflict, a program designed to stress the Rete matching algorithm. A methodology for characterizing knowledge-based system software and evaluating its performance is provided.
Two examples of dangerous rule interactions--redundancy and rule inconsistency--are reviewed. Their implementation is described in the context of the Defense Advanced Research projects Agency (DARPA) expert-system Val...
详细信息
Two examples of dangerous rule interactions--redundancy and rule inconsistency--are reviewed. Their implementation is described in the context of the Defense Advanced Research projects Agency (DARPA) expert-system Validation Associate (DEVA). Emphasis has been placed on devising strategies that can be used on atypical knowledge bases, i.e., those containing an especially hostile search space. A hostile search space is one which is very bushy and/or contains possibly many cycles in the rules. Techniques for detecting redundancy and rule-inconsistency anomalies in the absence of facts are discussed. Two approaches are considered: the restricted generate-and-test approach and residue analysis.
Production system implementations of expertsystems are becoming more prevalent in a large number of diverse specialties, but the relatively slow execution speed of these systems precludes their use in most realtime a...
详细信息
There are a series of expectations inherently built into the successful completion of a program or a project: the client's expectations for the delivery of services, the company's expectations of its employees...
详细信息
There are a series of expectations inherently built into the successful completion of a program or a project: the client's expectations for the delivery of services, the company's expectations of its employees' performance, and the expectations of the employees themselves for reward and recognition. It is suggested that a bridge of expectations to interface these success factors is a key ingredient to effective project management. A grading structure with four discrete levels, starting from entry level to expert, and covering every discipline required by its projects is described. The individual's accomplishments are verified through structured interviews and reference checks designed to respond to specific questions that have been developed. It precisely defines the job requirements, accurately screens and verifies the individual's competence to match the job requirements, and implements a standardized, equitable compensation system. On-the-job training and experience then allow individuals to advance themselves through a reclassification process to avail themselves of promotion. This grading structure is presently being used at multiple locations across the country.
This paper describes the Intel Touchstone Gamma Prototype, a distributed memory MIMD parallel computer based on the new Intel i860 floating point processor. With 128 nodes, this system has a theoretical peak performan...
详细信息
In this paper we examine the problem of parallel program synchronization in the distributed memory system, (DMS). We assume that the DMS is composed of several processing elements, (PE), interconnected by a subnet, (S...
详细信息
暂无评论