一种基于CBR的知识库求精模式

由于领域知识以及人们的认识进程具有进化的特性,领域模型总是不完备的,为了增强基于知识系统的自适应能力和可靠性,知识库求精已成为机译系统等基于知识系统实用化的必经阶段。本文给出了一些知识库求精原则;结合机器翻译知识库建构维护的实际需求,提出了一种基于CBR(Case-basedReasoning)的知识库求精模式。该求精模式以提高系统有效性为核心,以错误严重性为指示器,择重优     

MIKRS：用于知识库求精的模型推理系统

知识求精是知识获取的一个重要方面,本文主要介绍了知识库求精的一些概念、理论与方法,给出了在MIKRS系统中所实现的知识库调试和求精的思想及描述算法,并在文章的最后给出了系统运行的一个实例。     

基于伪自然语言理解的CAI开发平台

基于伪自然语言理解,提出并实现了一个种高效率的知识获取方法,并把它用一诉开发中。首先知识工程师利用来自然语言的ＢＬ语言 写书本自然描述,然后利用知识编译系统处理ＢＬ程序以高效率地实现书本知识获取,再后领域专家在书本知识库的基本语义呆引导下利用知识求精系统对书本知识库加以少许求精,接着对领域知识库动态全局规划,把领域知识分解成一个个概念,最后通过方法生成组织成一个个课文传授予学生。     

一个综合知识发现与知识求精系统--XFKDRS

介绍了一个综合知识发现与知识求精系统-XFKDRS.它由分类规则、关联规则、序贯模式、相似模式、聚类模式组成知识发现模块.知识求精首先在领域知识可视化的基础上,集成了基于遗传知识树、知识型人工神经网络和基于解释学习的求精方法.最后将新知识转换成雄风专家系统工具XF6.2的知识表达形式,添加到其知识库中,完成专家系统的自动知识获取.     

ESDDTL系统知识库的建立

本文介绍了ＥＳＤＤＴＬ知识库的构造方法。通过对知识整理，表示，获取，组织及求精等各个环节的设计与描述，完整地给出了一个医学诊疗专家系统知识库的建立过程。     

NNKBST：神经网络知识库系统支撑环境

为了有效地解决知识获取这一瓶颈问题，本文提出用神经网络来进行知识获取，以弥补传统归纳学习方法的不足，并以ＮＮ知识获取技术为核心，扩展为ＮＮ知识库系统支撑环境中一些全面技术的研究，包括ＮＮ知识表示和问题求解机制，ＮＮ知识库的推理，维护和求精机制等。     

一种专家系统知识获取时的属性约简算法

知识获取是构造专家系统的“瓶颈”,提供准确的推理知识是进行决策规划的关键。文中运用粗糙集理论,通过粗糙集的约简消除冗余的条件属性,实现对知识库的精简。首先研究知识获取,在阐明知识的层次结构基础上,给出了概念化、形式化、知识库求精三个知识获取过程;然后研究属性约简算法,在研究集合差异度和属性的重要性、约简算法推导过程的基础上,给出了属性约简算法的六个步骤。最后根据属性约简算法及其步骤,对功能点分析法构建软件成本估算专家系统时,组成技术复杂因子的14个因素进行了约简。     

知识求精

本文讨论了知识系统的一般建造模型。重点分析了知识系统开发过程中的求精阶段。提出了三级知识求精模式。探讨了知识求精的有关问题及发展趋势。     

基于神经网络结构学习的知识求精方法

知识求精是知识获取中必不可少的步骤．已有的用于知识求精的ＫＢＡＮＮ（ｋｎｏｗ ｌｅｄｇｅ ｂａｓｅｄ ａｒｔｉｆｉｃｉａｌｎｅｕｒａｌｎｅｔｗ ｏｒｋ）方法,主要局限性是训练时不能改变网络的拓扑结构．文中提出了一种基于神经网络结构学习的知识求精方法,首先将一组规则集转化为初始神经网络,然后用训练样本和结构学习算法训练初始神经网络,并提取求精的规则知识．网络拓扑结构的改变是通过训练时采用基于动态增加隐含节点和网络删除的结构学习算法实现的．大量实例表明该方法是有效的     

专家系统的知识求精技术

本文由两部分组成.第一部分给出一个知识求精系统的一般模式。第二部分详细介绍几个著名的专家系统知识求精技术,并给出简短的评论。最后,提出一些值得进一步研究的问题。     

A NEW APPROACH FOR THE INCREMENTAL DEVELOPMENT OF RETRIEVAL FUNCTIONS FOR CBR

We present a new approach to the effective development of complex retrieval components for case-based reasoning systems (CBR). Our approach goes beyond the traditional CBR approach by allowing an incremental refinement of an existing retrieval knowledge base during routine use of the system. The refinement takes place through a direct expert-system interaction while the expert is accomplishing their given tasks. We lend ideas from ripple-down rules (RDR), a proven method for the very effective and efficient acquisition of classification knowledge during the routine use of a knowledge-based system (KBS).

In our approach the expert is only required to provide explanations of why, for a given problem, a certain case should be retrieved. Incrementally a complex retrieval knowledge base as a composition of many simple retrieval functions is developed. This approach is effective with respect to both the development of highly tailored and complex retrieval knowledge bases for CBR as well as providing an intuitive and feasible approach for the expert. The approach has been implemented in our CBR system MIKAS (Menu construction using an Incremental Knowledge Acquisition System) that allows to automatically construct a menu that is strongly tailored to the individual requirements and food preferences of a client.     

VALIDATION OF FIRST-ORDER RULE-BASED SYSTEMS1

Knowledge base validation and knowledge base refinement aim to help the expert to improve an existing knowledge base. They deal with the final knowledge acquisition phase and rely on a quality measurement of an existing knowledge base. We present our approach to knowledge base refinement, which is based on results in the domain of knowledge base validation. Our approach is based on a general consistency definition of a knowledge base and on a study of causes of knowledge base inconsistency. Our approach relies significantly on a differentiation of sure and expert knowledge in the knowledge base. We have implemented a system that has two phases: one computational phase decides on the consistency of a knowledge base, and, if necessary, a second phase helps the expert to interactively update the knowledge base. We present some related work in the domain. We illustrate the use of our system with an example.     

Refining Numerical Constants in First Order Logic Theories

This paper proposes a method for refining numerical constants occurring in rules of a knowledge base expressed in a first order logic language. The method consists in tuning numerical parameters by performing error gradient descent. The knowledge base to be refined can be manually handcrafted or automatically acquired by a symbolic relational learner, able to deal with numerical features. The results of an experimental analysis performed on four case studies show that the refinement step can be effective in improving classification performances.     

An interactive tool for knowledge base refinement

We present a tool that combines two main trends of knowledge base refinement. The first is the construction of interactive knowledge acquisition tools and the second is the development of machine learning methods that automate this procedure. The tool presented here is interactive and gives experts the ability to evaluate an expert system and provide their own diagnoses on specific problems, when the expert system behaves erroneously. We also present a database scheme that supports the collection of specific instances. The second aspect of the tool is that knowledge base refinement and machine learning methods can be applied to the database, in order to automate the procedure refining the knowledge base. In this paper we examine the application of inductive learning algorithms within the proposed framework. Our main goal is to encourage the experts to evaluate expert systems and to introduce new knowledge, based on their experience.     

Formal derivation of rule-based programs

It is shown that a combination of specification and program refinement may be applied to deriving efficient concurrent rule-based programs. Specification refinement is used to generate an initial rule-based program that is refined into a program which is highly concurrent and efficient. This program derivation strategy is divided into two major tasks. The first task relies on specification refinement. Techniques similar to those employed in the derivation of UNITY programs are used to produce a correct rule-based program having a static knowledge base. The second task involves program refinement and is specific to the development of concurrent rule-based programs. It relies heavily on the availability of a computational model, such as Swarm, that has the ability to dynamically restructure the knowledge base. The ways in which a Swarm program can be translated to OPS5 specifically, given some restrictions, while maintaining the correctness criteria are discussed     

Condition state evaluation of existing reinforced concrete bridges using neuro-fuzzy hybrid system

This article presents a new approach for developing a concrete bridge rating expert system for deteriorated concrete bridges, constructed from multi-layer neural networks. The system evaluates the performance of concrete bridges on the basis of a simple visual inspection and technical specifications. The main reason of applying the neural network is that it performs fuzzy inference in the network, facilitates refinement of the knowledge base by use of the back-propagation method, and prevents not only the inference mechanism of the expert system but also the knowledge base after machine learning from becoming a black box.     

Kave: a tool for knowledge acquisition to support artificial ventilation

A decision support system for artificial ventilation is being developed. One of the fundamental goals for this system is the application of the system when a domain expert is not present. Such a system requires a rich knowledge base. The knowledge acquisition process is often considered to be the bottleneck in acquiring such a complete knowledge base. Since no single available method, for example interviewing domain experts, is sufficient for removing this bottleneck, we have chosen a combination of different methods. The different backgrounds of knowledge engineers and domain experts could cause communication restrictions and difficulties between them, e.g. they might not understand each others knowledge domain and this will affect formulation of the knowledge. To solve this problem we needed a tool which supports both the knowledge engineer and the domain expert already from the initial phase of developing the knowledge base. We have developed a knowledge acquisition system called KAVE to elicit knowledge from domain experts and storing it in the knowledge base. KAVE is based on a domain specific conceptual model which is a result of cooperation between knowledge engineers and domain experts during identification, design and structuring of knowledge for this domain. KAVE includes a patient simulator to help validate knowledge in the knowledge base and a knowledge editor to facilitate refinement and maintenance of the knowledge base.     

基于数据库的燃煤锅炉事故诊断专家系统的设计与实现

采用MS SQL Server7．0设计知识库，Visual Basic 6．0编程实现了燃煤锅炉事故诊断专家系统。本系统知识表示采用了基于概率逻辑的产生式规则形式，并用数据库的方法存储管理知识库。推理机采用规则值的方法，并应用主观Bayes理论建立了不确定性推理模型。实现了一种较为理想的不精确推理。