一种面向对象的专家系统设计方法

近年来,已经有不少面向对象技术运用于专家系统的研究。本文提出一种新的面向对象专家系统的设计方法。首先介绍系统总体结构,给出知识表示模型和推理机制,并且引入RMG的概念,进一步提出运用RMG完成推理过程的算法。最后把提出的方法应用于中西医结合糖尿病诊疗专家系统的开发。实践证明此方法是有效的。     

面向对象的知识表示方法

本文在分析了传统的知识表示方法基础上，着重探讨了面向对象的知识表示的方法，特点，推进及程序实现问题。     

面向对象的知识表示与混合专家系统设计

面向对象的知识表示是近些年来才提出的一种知识表示方法，本文探讨了它的基本原理，并利用它来构筑知识库以及实现混合专家系统工具。文中着重讨论了面向对象的知识表示，面向对象的知识库和混合专家系统工具。     

结合面向对象技术的知识表示系统——OOplog

介绍了一个新的结合面向对象技术的知识表示系统ＯＯｐｌｏｇ。以前的一些系统都是建立在Ｐｒｏｌｏｇ解释器的基础上，间接实现的，效率较低。而ＯＯｐｌｏｇ系统将面向对象技术与逻辑语言相结合，采用Ｃ语言直接实现，所以效率较高。ＯＯｐｏｌｇ具有较强的建模能力，并提供了统一的推理机。     

面向对象知识表示在起重机专家系统中的应用

知识表示是专家系统的关键组成部分。在对知识表示的方法进行探究的基础上,根据起重机操作专家系统的特点,采用面向对象与产生式表示方法相结合的混合表示方法对其专家知识进行表示,并在C＋＋程序语言环境下实现起童机操作知识的表示,同时给出起重机操作的知识表示实例,用此法开发的专家系统具有代码短、易维护、易修改和易扩充的优点。阐述了起重机操作专家系统如何用面向对象方法封装规则、建立知识库、实现知识推理和具体的应用要点。     

面向对象知识表示在起重机专家系统中的应用

知识表示是专家系统的关键组成部分.在对知识表示的方法进行探究的基础上,根据起重机操作专家系统的特点,采用面向对象与产生式表示方法相结合的混合表示方法对其专家知识进行表示,并在C 程序语言环境下实现起重机操作知识的表示,同时给出起重机操作的知识表示实例,用此法开发的专家系统具有代码短、易维护、易修改和易扩充的优点.阐述了起重机操作专家系统如何用面向对象方法封装规则、建立知识库、实现知识推理和具体的应用要点.     

面向对象模型对复杂结构知识的表示方法

传统关系型数据库在表示专家系统中具有嵌套、递归等复杂结构的知识时遇到了困难。对此，本文提出了一种采用对象模型的解决方案。该方案在关系数据库的基础上使用映射原理实现了一种面向对象的模型，利用该模型可以较好地在数据库中表示各类复杂结构的知识。该方案已经在地震预报专家系统中得到应用，并取得了较好的效果。     

面向对象表示法在火控系统故障诊断知识表示中的应用

通过与目前几种常用的知识表示方法的优缺点的对比，提出了火控系统故障诊断知识的一种新的表示方法——面向对象表示法。阐述了面向对象知识推理的特点和具体的应用。     

面向对象的工程数据库模型

提高数据库模型的表达能力是增强数据库系统的性能与功能的关键。本文基于面向对象的思想，吸收语义数据模型、知识的演绎推理机制和图形的组成关系运算，提出了一个面向对象的工程数据库模型。讨论了面向对象的特征、语义构造能力，介绍了知识的表示和图形的生成方法。     

基于问题的面向对象知识表示方法的研究

基于问题的面向对象知识的表示方法,克服了构建知识库过程中对知识进行形式化的繁杂的过程和以纯文本作为检索对象效率低的缺点.这种方法以问题类别作为属性的标记,并标注在原始 文本中,利用已标记的文本作为知识库的组成元素,关系数据库储存对象的类属关系.在以文本作为知识库元素的问答系统中用基于问题的面向对象知识表示方法,使得通过对问话的分析就能够迅速地找到答案的位置,提高了直接从原始文本中进行信息提取的效率.     

Constraint satisfaction with an object-oriented knowledge representation language

This article gives a detailed presentation of constraint satisfaction in the hybrid LAURE language. LAURE is an object-oriented language for Artificial Intelligence (AI) applications that allows the user to combine rules, constraints, and methods that cooperate on the same objects in the same program. We illustrate why this extensibility is necessary to solve some large and difficult problems by presenting a real-life application of LAURE. We describe the syntax and the various modes in which constraints may be used, as well as the tools that are proposed by LAURE to extend constraint resolution. The resolution strategy as well as some implementation details are given to explain how we obtain good performances.     

An integrated system for knowledge sharing among heterogeneous knowledge derivation systems

Solving problems in a complex application domain often requires a seamles integration of some existing knowledge derivation systems which have been independently developed for solving subproblems using different inferencing schemes. This paper presents the design and implementation of an Integrated Knowledge Derivation System (IKDS) which allows the user to query against a global database containing data derivable by the rules and constraints of a number of cooperative heterogeneous systems. The global knowledge representation scheme, the global knowledge manipulation language and the global knowledge processing mechanism of IKDS are described in detail. For global knowledge representation, the dynamic aspects of knowledge such as derivational relationships and restrictive dependencies among data items are modeled by a Function Graph to uniformly represent the capabilities (or knowledge) of the rule-based systems, while the usual static aspects such as data items and their structural interrelationships are modeled by an object-oriented model. For knowledge manipulation, three types of high-level, exploratory queries are introduced to allow the user to query the global knowledge base. For deriving the best global answers for queries, the global knowledge processing mechanism allows the rules and constraints in different component systems to be indiscriminately exploited despite the incompatibilities in their inferencing mechanisms and interpretation schemes. Several key algorithms required for the knowledge processing mechanism are described in this paper. The main advantage of this integration approach is that rules and constraints can in effect be shared among heterogeneous rule-based systems so that they can freely exchange their data and operate as parts of a single system. IKDS achieves the integration at the rule level instead of at the system level. It has been implemented in C running in a network of heterogenous component systems which contain three independently developed expert systems with different rule formats and inferencing mechanisms.Database Systems Research and Development Center, Department of Computer Information Sciences, Department of Electrical Engineering, University of Florida     

An object-oriented model for FMS control

The flexible manufacturing system (FMS) is a distributed network of heterogeneous programmable manufacturing machinery, such as assembly lines and numerically controlled machines. Despite these interconnected, programmable hardware elements, the success of building a truly flexible manufacturing system has been limited so far, owing to the lack of flexibility in its control software layer. In integrating heterogeneous machinery, many existing FMS control software systems depend structurally on specific machinery and job-scheduling strategies, and thus it is difficult to incorporate new developments in FMS organization and operational requirements. In searching for an open architecture for the FMS control software system, this paper presents an object-oriented FMS data model. Among others, it represents each physical cluster of related machinery (called a flexible manufacturing cell) as an object. To facilitate the integration of heterogeneous physical cells, such cell objects share a common protocol of interacting with the main control process through inheritance from the abstract cell class. Other related physical and abstract entities in FMS are also modelled as objects, with their similarity and difference captured in inheritance hierarchies. To verify the proposed approach experimentally, a prototype FMS control software system named FREE (FMS Runtime Executive Environment) has been implemented on top of a commercial object-oriented database system.     

知识库系统中语义网知识的表示

基于对知识库系统中知识表示的理解,提出了在语义网环境下知识库系统的建立和相应的知识表示,并将此扩展到语义网中整个的知识表示系统。     

An Object-Oriented Transaction Model

In object-oriented database systems(OOBSs),the traditional transaction models are no longer suitable because of the difference between the object-oriented data model(OODM)and the conventional data models(e.g.relational data model).In this paper,transction models for advanced database applications are reviewed and their shortcomings are analyzed.Exchangeability of operations is proposed instead of commuativity and recoverability for using more semantics in transaction management.As a result,an object-oriented transaction model(in short,OOTM)is presented.It is not modeled for some special application,but directly based on object-oriented paradigms.A transaction is regarded as an interpretation of a metho.Each transaction(even subtransactions)keeps relative ACID(Atomicity,Consistency,Isolation,Durability)properties,therefore the special problems appearing in OOBSs such as“long transactions”,“visibility of inconsistent database state”can be solved.     

面向个性化推荐的node2vec-side融合知识表示

推荐系统中知识图谱对系统的推荐效果起到很重要的作用,图谱中的知识表示成为影响推荐系统的关键因素,这也成为当前的研究热点之一。针对推荐系统中知识图谱的结构特点,在传统node2vec模型基础上增加关系表示和多元化游走策略,提出一种基于node2vec的知识表示node2vec-side,结合推荐系统知识图谱网络结构,旨在挖掘大规模推荐实体节点间潜在的关联关系,降低表示方式复杂度,提高可解释性。经过时间复杂度分析可知,提出的知识表示方式在复杂度上低于Trans系列和RGCN。在传统知识图谱数据集FB15K、WN18和推荐领域数据集MovieLens-1M、Book-Crossing、Last.FM上分别进行链接预测对比实验。实验结果表明,在MovieLens-1M数据集上,Hits@10分别提升了5.5%～12.1%,MRR提升了0.09～0.24;在Book-Crossing数据集上,Hits@10分别提升了3.5%～20.6%,MRR平均提升了0.04～0.24;而在Last.FM数据集上,hits@1提升了0.3%～8.5%,MRR平均提升了0.04～0.16,优于现有算法,验证了所提方法的有效性。     

An object-oriented knowledge representation for spatial information

An abstract formalism for the representation of spatial knowledge is suggested. The focus is on the development of a comprehensive representation scheme for pictorial information in which the knowledge model of the given world has a high degree of perceptual similarity to a typical user's view of the same world. The model that has been developed uses the object-oriented method of knowledge representation. The intention is that with this model any user of the system will be equipped to depict pictorial information easily and will be able to portray spatial as well as conceptual abstractions, generalizations, and rules at various levels     

Model for knowledge and legal expert systems

This paper presents a four layer model for working with legal knowledge in expert systems. It distinguishes five sources of knowledge. Four contain basic legal knowledge found in published and unpublished sources. The fifth consists of legal metaknowledge. In the model the four basic legal knowledge sources are placed at the lowest level. The metaknowledge is placed at levels above the other four knowledge sources. The assumption is that the knowledge is represented only once. The use of metaknowledge at various levels should make it possible to use the appropriate knowledge for the problem presented to the system. The knowledge has to be represented as closely to the original format as possible for this purpose. Suitable representation formalisms for the various types of knowledge in the five knowledge sources are discussed. It is not possible to indicate a best representation formalism for each knowledge source.