摘要:

Production systems are still regarded as one of the most important paradigms for solving complex tasks. However, in embedded control systems production systems have not yet been applied intensively. We identify open issues for production systems in process control by exemplary presenting two industrial fielded expert systems. We describe how these expert systems have been applied to process control, discuss lessons learned, and show how practical experience has influenced our own research activities. Finally we identify future research directions of general importance.

ISSN:0883-9514    

DOI:10.1080/08839519308949992

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

Although successful in medical diagnostic problems, inductive learning systems were not widely accepted in medical practice. In this paper two different approaches to machine learning in medical applications are compared: the system for inductive learning of decision trees Assistant, and the naive Bayesian classifier. Both methodologies were tested in four medical diagnostic problems: localization of primary tumor, prognostics of recurrence of breast cancer, diagnosis of thyroid diseases, and rheumatology. The accuracy of automatically acquired diagnostic knowledge from stored data records is compared, and the interpretation of the knowledge and the explanation ability of the classification process of each system is discussed. Surprisingly, the naive Bayesian classifier is superior to Assistant in classification accuracy and explanation ability, while the interpretation of the acquired knowledge seems to be equally valuable. In addition, two extensions to naive Bayesian classifier are briefly described: dealing with continuous attributes, and discovering the dependencies among attributes.

ISSN:0883-9514    

DOI:10.1080/08839519308949993

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

The robustness of NLP techniques can be improved by the use of “shallow” methods such as statistical analysis in combination with traditional knowledge-based methods, such as syntax and semantics

ISSN:0883-9514    

DOI:10.1080/08839519308949994

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

A correlation-based approach to automatic face recognition requires adequate normalization techniques. If the positioning of the face in the image is accurate, the need for shifting to obtain the best matching between the unknown subject and a template is drastically reduced, with considerable advantages in computing costs. In this paper, a novel technique is presented based on a very efficient eyes localization algorithm. The technique has been implemented as part of the “electronic librarian” of MAIA, the experimental platform of the integrated Al project under development at IRST. Preliminary experimental results on a set of 220 facial images of 55 people disclose excellent recognition rates and processing speed.

ISSN:0883-9514    

DOI:10.1080/08839519308949995

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

This paper is a constructive critique of the BOXES methodology, based on more than fifteen years of study by the author, and is designed to be an evaluative reference in regard to its future application in real-time, real world systems. The paradigm is examined in some detail with particular reference to sections that convey the “spirit of the method” rather than paying too much attention to the mathematical processes involved, or to any particular application. Areas of concern are discussed, and alterations to some of the basic algorithmic components are suggested. The paper concludes with a “futures” section that highlights further studies in the application of this truly BLACK BOX controller.

ISSN:0883-9514    

DOI:10.1080/08839519308949996

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

The diagnostic task is a problem in which one is given a set of symptoms or manifestations observed from a diagnosed system and is required to find a parsimonious set of disorders in a given layer of the system. These disorders may produce (or explain) directly or indirectly (e.g., through disorder propagation) those symptoms by using his knowledge about the domain. The diagnostic problem solving for a complex system is proved to be particularly difficult because of the lack of efficient techniques on knowledge organization and knowledge integration at present for the problem. In this paper we address issues of expertise models and present a novel strategy for integrating multiple knowledge sources in the diagnostic problem solving of complex systems. More importantly, a Knowledge-Based System for Engine Diagnosis (KBSED) is introduced in this article. The KBSED is a knowledge-based information-processing system, integrating fault diagnosis, state detection, and recognition, with repair consultation for automobile engines without whole-block disassembly. An implementation architecture for this system is described, including hierarchical diagnosis, knowledge exploitation, knowledge control structure, and other techniques related to the diagnosis. Finally, the system structure and function of KBSED is described in detail.

ISSN:0883-9514    

DOI:10.1080/08839519308949997

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor

摘要:

This paper reviews and extends a formal model of multiagent symbol systems that define emergent symbolic behavior as a result of collective activities of very simple purely reactive components of the system. The formal model—the colony of grammars—is conceptually and technically rooted in the theory of formal grammars and languages (especially in the theory of grammar systems). This paper draws some parallels and sketches some principal differences between the symbol system model and real societies of behavior-based robotic systems of purely reactive components.

ISSN:0883-9514    

DOI:10.1080/08839519308949998

出版商:Taylor & Francis Group    

年代:1993

数据来源: Taylor