Reported by the author is an expert system which enhances the capability of a nonlinear optimization program. Included are discussions on the knowledge acquisition aspect and the advantageous use of neural networks to...
详细信息
Reported by the author is an expert system which enhances the capability of a nonlinear optimization program. Included are discussions on the knowledge acquisition aspect and the advantageous use of neural networks to accomplish knowledge acquisition. The expert system diagnoses the output of the NPSOL optimization package and seeks to find errors in the user-supplied inputs. These errors result from poor modeling formulation of the problem and poor parameter settings when running the program. A prototype system is installed on a Unix machine.< >
Although several methods for economic justification of general information system and office automation system are well documented, research on the feasibility of using these methods in justifying intelligentsystems ...
详细信息
Although several methods for economic justification of general information system and office automation system are well documented, research on the feasibility of using these methods in justifying intelligentsystems and on particular methods for intelligentsystem justification are very limited. This paper proposes a process approach to economic justification of intelligentsystem. As examples of that process approach, a method based on queuing models for economic justification of intelligentsystem application in service delivery and an activity price model for justification of intelligentsystem application in managerial decision support are discussed in detail.< >
SkillSet, a hypertext intelligentsystem that provides advice and suggests management tools appropriate for solving problems encountered in project management contexts, is described. The suggestions are drawn from a d...
详细信息
The integration of neural networks and expert systems creates the potential for systems that are more powerful than ones using either of the techniques alone. The project presented uses hybrid systems for analyzing co...
详细信息
The integration of neural networks and expert systems creates the potential for systems that are more powerful than ones using either of the techniques alone. The project presented uses hybrid systems for analyzing corporate databases. In this amplification, a neural network is trained to recognize patterns of information in an employee database, and an expert system component combines those results with a rule-based analysis to recommend staffing for projects. Results are presented along with implementation details and recommendations for further work in this area.< >
This paper describes the methodology and how the authors developed it for eliciting potential expert systemprojects at the Internal Revenue Service (IRS). They intertwined demonstrations of existing IRS expert system...
详细信息
This paper describes the methodology and how the authors developed it for eliciting potential expert systemprojects at the Internal Revenue Service (IRS). They intertwined demonstrations of existing IRS expert systems and examples from non-IRS applications with questions like 'Do you have anything like this?' to stimulate ideas of similar tasks that might be amenable to solution via an expert system. They also interspersed minitalks on artificial intelligence (AI) terminology at the appropriate level when they felt that such use could not be avoided. The result in one division of the IRS was over 20 potential future expert systems.< >
The development of a prototype system for accessing lessons learned to support military planning is described. The system accesses lessons learned, developed in various formats for a variety of automated systems, in t...
详细信息
The development of a prototype system for accessing lessons learned to support military planning is described. The system accesses lessons learned, developed in various formats for a variety of automated systems, in the relevant planning context. The requirements-definition process, implementation, and observations on system development are discussed. Representation and indexing schemes for lessons learned are described. Although primarily used in defense-related areas, the lessons-learned system is also applicable to nondefense areas (e.g., environmental management, energy facility operation).< >
Educational Research in Data Abstractions (ERDA) is an intelligent tutorial system. The system's focus is to present animated and graphical representations of data abstractions and algorithms to supplement CS2 lec...
详细信息
Educational Research in Data Abstractions (ERDA) is an intelligent tutorial system. The system's focus is to present animated and graphical representations of data abstractions and algorithms to supplement CS2 lectures. The authors deal with the research subjects of ERDA which include the following topics: a user-centered, highly portable interface that provides the user with a persistent framework and the ability to initiate nonlinear exploration; object-oriented lesson development techniques; standardization and documentation of lesson features that provides consistency and ensures future portability; and emphasis on interactive, user-directed learning.
How can one identify the complexity of a system? Having identified it, how can one cope with it? Facing these heavy issues, the authors explain and sketch their thoughts by stating the necessity of building an intelli...
详细信息
How can one identify the complexity of a system? Having identified it, how can one cope with it? Facing these heavy issues, the authors explain and sketch their thoughts by stating the necessity of building an intelligentsystem to enable them to solve or partly solve the problems by dealing with one aspect of the complexity-the part derived from systems' interaction. 'Coping with the complexity' then means to bring different levels of a systems' interaction onto one so called 'manageable level'. The authors propose the development of an intelligentsystem on the following levels: a functional level including a scope of its functions; a systematic level including a set of principles for its integration; a modelling level including an approach to model it.< >
The work presented is part of a more important project which aims at designing federated knowledge based-systems (FKBS). The purpose of FKBSs is to make autonomous, heterogeneous and eventually geographically distribu...
详细信息
The work presented is part of a more important project which aims at designing federated knowledge based-systems (FKBS). The purpose of FKBSs is to make autonomous, heterogeneous and eventually geographically distributed KBSs cooperate with each other. The first step towards such systems consists in facilitating the designer's access to the pre-existing expert processing capabilities of the organization. In that regard, the purpose of the authors is to propose a deductive database (DDB) which one can record, structure, maintain and produce information about the existing KBS and their reusable components (knowledge base, control process etc.). In addition to intelligentsystemprojects support, the DDB can also serve for knowledge communication, human training and software re-usability purposes.< >
In industry, engineers have to apply their knowledge and understanding to real problems. They need to develop a qualitative understanding of how industrial systems work. The authors discuss the possibility of integrat...
详细信息
In industry, engineers have to apply their knowledge and understanding to real problems. They need to develop a qualitative understanding of how industrial systems work. The authors discuss the possibility of integrating simulation and artificial intelligence techniques in industry and particularly in educational applications. Then they describe an architecture of an industrial educational system taking into account both simulation and artificial intelligence. In this context, the industrial proceeding is decomposed in basic phases; and small comprehensive elements (knowledge elements) are defined. Many lessons (dialogues) and educational tools are integrated. The tutor chooses admissible lessons and filters only one among them to be proposed to the student eliminating the ones which deviate from the educational strategy.< >
暂无评论